Skip to Content
  • 3.4
    Impact Factor
  • 9.0
    CiteScore
  • 22 days
    Submission to First Decision
  • 5 days
    Acceptance to Publication

Applied System Innovation

Applied System Innovation (ASI) is an international, peer-reviewed, open access journal on integrated engineering and technology, published monthly online. It is the official journal of the International Institute of Knowledge Innovation and Invention (IIKII). 

Get Alerted

Add your email address to receive forthcoming issues of this journal.

All Articles (1,058)

  • Article
  • Open Access

Industry 5.0 calls for decision-support systems that are human-centric, resilient, and transparent. However, in bulk-material logistics, predictive models are often developed as static tools, with limited attention to process structure, interpretability, and changing operating conditions. This study proposes a decision-support framework that combines the DMAIC cycle, machine learning (ML), explainable artificial intelligence (XAI), and human-in-the-loop model management, and applies it to rail unloading at a phosphate-handling facility. DMAIC was first used to structure process knowledge and identify the main drivers of daily throughput, including train frequency, wagon filling conditions, unloading duration, and operational losses. ML models were then evaluated for early prediction of final daily throughput, using only information available after the first train, with and without recent historical context. Adding historical features reduced validation MAE across all four tabular algorithms, with V2-XGBoost reaching a validation MAE of 2043 T/J and an RMSE of 2368 T/J. On the final chronological test period, however, all fixed ML models returned negative R2 values. Between the training and test periods, mean throughput increased from 19,070 to 22,233 T/J, and the proportion of seven-train days rose from 16.5% to 68.1%, indicating a marked regime shift. SHAP analysis was used to explain model behavior at the global and local levels, including regime-dependent predictive relationships. Based on these findings, we outline a human-supervised architecture linking early prediction, explanation, performance monitoring, and model review, in which operators and managers retain final decision authority. The study highlights the importance of evaluating predictive reliability under changing operating conditions in industrial logistics.

Appl. Syst. Innov.

20 November 2026

Representative view of the industrial rail-unloading environment.
  • Feature Paper
  • Article
  • Open Access

This study compares twenty-one machine-learning classifiers for the automatic classification of cardiac abnormalities from single-channel ECG signals, with particular attention to predictive performance and computational requirements relevant to connected cardiac monitoring devices. Five heartbeat classes are considered: normal beat (NOR), atrial premature contraction (APC), left bundle branch block (LBBB), right bundle branch block (RBBB), and ventricular premature contraction (VPC). ECG signals were extracted from the MIT-BIH Arrhythmia Database and segmented into one-second windows centered on annotated beat positions. The preprocessing pipeline included wavelet-based denoising, normalization, and class balancing using a hybrid RandomUnderSampler–SMOTE strategy. All twenty-one classifiers were evaluated using stratified five-fold cross-validation on the training partition, while an independent held-out test set was preserved for complementary evaluation. Performance was assessed using accuracy, macro precision, macro recall, macro F1-score, macro-AUROC, training time, and inference latency. Extra Trees achieved the highest mean cross-validation accuracy (98.12±0.16%) and macro F1-score (95.05±0.52%), together with the best average rank across folds. Random Forest, XGBoost, LightGBM, and SVM also achieved strong predictive performance. The Friedman test revealed significant global differences among the evaluated classifiers (χ2=98.0468, ), whereas the subsequent Nemenyi analysis indicated that the leading classifiers were statistically comparable at the 0.05 significance level. Considering predictive performance, stability across cross-validation folds, and computational cost, Extra Trees and LightGBM were retained as promising candidates for further investigation in resource-constrained ECG monitoring applications. These results emphasize that classifier selection for future embedded biomedical systems should consider not only predictive performance, but also statistical stability and computational requirements.

Appl. Syst. Innov.

20 November 2026

Example of the ECG preprocessing pipeline applied to MIT-BIH Arrhythmia Database (v1.0.0) record 101, lead MLII, over the interval 0–720 samples (0–2 s at 360 Hz).
  • Article
  • Open Access

Mathematical Analysis of Non-Steady-State Enzymatic Biofuel Cells: A Planar Model Generalized to Other Geometries

  • Daniel Samuel,
  • Mallikarjuna Mohanasundaraganesan and
  • Selvam Arunachalam
  • + 1 author

This study develops approximate, closed-form (semi-analytical) expressions for the time-dependent evolution of substrate and mediator concentration profiles and the resulting transient current response in an enzymatic biofuel cell electrode. The analysis is developed for a planar catalyst film, which is the physically rigorous configuration, and is then carried over to cylindrical and spherical shapes through a generalized shape-factor ( ) formulation; the curved-geometry results are presented as a formal mathematical extension of the planar model rather than as literal models of wire- or bead-electrode devices. The coupled nonlinear ping-pong kinetics are treated with the Laplace Homotopy Perturbation Method, in which the Michaelis–Menten consumption term is represented by linearized reaction moduli; the resulting closed-form expressions are interpretable, and their accuracy is validated directly against the numerical solution of the full nonlinear equations over the operating range considered. Capturing electrode behavior during start-up from an initially unloaded film held at a fixed applied (Nernstian) potential, the model enables a systematic parametric sensitivity analysis of how the Thiele modulus, the Michaelis–Menten constants for substrate and mediator, the applied potential amplitude, the initial film conditions, and the generalized shape factor govern not only steady-state performance but also the response time and the monotonic transient approach to steady-state operation (with no current overshoot), going beyond prior treatments largely confined to the steady-state limit, and thereby offering mechanistic insight and design guidance for biofuel cell operation. The closed-form approximation reproduces the full nonlinear numerical solution to within about 1% for the planar electrode over ϕ≲2 at all times and to within 1–2% for the curved (formal) shape-factor extension for τ≳0.5, which delimits its domain of validity.

Appl. Syst. Innov.

18 November 2026

Schematic of the three bioanode configurations: (a) a planar enzyme–mediator film of thickness l on a flat electrode; (b) a cylindrical film coating a wire or fiber; and (c) a spherical film coating a particle. In every case, the electrode surface is at 
  
    x
    =
    0
  
 and the film/solution interface at 
  
    x
    =
    l
  
. Substrate diffuses inward from the bulk solution, and the mediator shuttles electrons to the electrode. The three geometries are analyzed within a single generalized coordinate framework, the curvature being represented by the shape factor N (
  
    N
    =
    0
  
, 1, 2).
  • Article
  • Open Access

Precision machining of tungsten carbide (WC) presents challenges in surface quality, dimensional accuracy, and processing efficiency. This study developed an integrated computer numerical control (CNC) machining workflow combining ultrasonic-assisted machining (USM), stage-specific parameter allocation, and localized helical finishing. Conventional machining and USM were compared using DR14C, followed by machining parameter adjustments and toolpath optimization. The selected workflow was subsequently applied to TR15C and KG5. Under identical baseline conditions, USM reduced surface roughness from Ra 0.600 to 0.510 μm on the flat surface and from 0.617 to 0.504 μm in the groove, representing reductions of 15.0% and 18.3%, respectively. Confocal measurements further revealed reductions in mean Ra from 0.255 to 0.133 μm and mean Rz from 2.208 to 1.560 μm. The optimized process reduced CNC controller-simulated machining time by 82.7%. Localized helical finishing achieved Ra values of 0.352 and 0.429 μm on the flat and groove surfaces, respectively. Scanning electron microscopy (SEM), backscattered electron (BSE) imaging, and energy-dispersive X-ray spectroscopy (EDS) provided supplementary surface and elemental characterization. These findings demonstrated the effectiveness of the integrated workflow in improving machining quality and controller-predicted efficiency.

Appl. Syst. Innov.

17 November 2026

Experimental Procedure.

Featured Articles of Last Quarter

Sets of operating modes, technical conditions, and monitoring parameters and their interrelation.
Flowchart of the literature selection process and methodological approach.

Highly Accessed Articles

News & Conferences

Latest Issues

Open for Submission

Journal Sections

XFacebookLinkedIn
Appl. Syst. Innov. - ISSN 2571-5577