Bermawi Priyatna Iskandar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Bermawi Priyatna Iskandar has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Hennie Husniah
Department of Industrial Engineering, Faculty of Engineering, Universitas Langlangbuana, Bandung 40261, Indonesia
3 shared publications Andi Cakravastia
Inst Teknol Bandung, Fac Ind Engn, Ind Engn, Jalan Ganesha 10, Bandung 40132, Indonesia
2 shared publications- Andi Cakravastia
Institut Teknologi Bandung
2 shared publications R Wangsaputra
Department of Industrial Engineering, Bandung Institute of Technology Ganesha 10, Bandung, 40132, Indonesia
2 shared publicationsA Cakravastia
Department of Industrial Engineering, Bandung Institute of Technology Ganesha 10, Bandung, 40132, Indonesia
1 shared publicationAndreas Tjendra
School of Computing Technologies, RMIT University, Melbourne, VIC 3001, Australia
1 shared publication
Latest Publications
- Open Access
- 1 August 2026
- Published by Ital Publication in Emerging Science Journal
- Vol. 10 (4) , 2032-2060
Fault Signature Maps: A Signal-Level Explainable Artificial Intelligence Framework for Bearing Fault Diagnosis
Deep learning models for bearing fault diagnosis on the Case Western Reserve University (CWRU) dataset routinely achieve near-perfect accuracy. Yet their decisions remain largely opaque at the signal level, so engineers cannot determine where in the raw vibration waveform the model focuses. This study aims to bridge the interpretability gap in deep learning-based bearing fault diagnosis by developing a signal-level explainability framework for the Wide Deep Convolutional Neural Network (WDCNN) on the Case Western Reserve University (CWRU) dataset. SHAP DeepExplainer is applied directly to raw 2,048-point vibration segments to produce per-sample-point attribution maps; Fault Signature Maps (FSMs) are formalized as class-averaged SHAP fingerprints in three variants (signed, absolute, and variance) and validated via discriminability index, severity monotonicity, and split-half stability, complemented by a three-variant ablation study examining how architectural decisions affect both accuracy and explainability. On the 10-class CWRU dataset under a rigorous temporal split (1,240/310/750 samples), WDCNN achieves 99.87% test accuracy with a macro F1-score of 0.997. FSMs demonstrate high reproducibility (split-half stability = 0.940); severity monotonicity ranges from 8.6% to 17.6%; and ablation reveals that removing Batch Normalization increases FSM discriminability by 2.3× at a 3.6% accuracy cost. As the first work applying SHAP DeepExplainer at full 2,048-point raw signal resolution for WDCNN, this study establishes that fault discrimination relies on transient impulse morphology rather than bearing characteristic frequencies, a finding invisible to feature-level XAI, and introduces a multi-resolution diagnostic framework bridging deep learning accuracy with physically interpretable vibration analysis for trustworthy deployment in safety-critical industrial environments.
- 1 August 2026
- Published by Ital Publication in Emerging Science Journal
- Vol. 10 (4) , 2032-2060
Open AccessFault Signature Maps: A Signal-Level Explainable Artificial Intelligence Framework for Bearing Fault Diagnosis
Deep learning models for bearing fault diagnosis on the Case Western Reserve University (CWRU) dataset routinely achieve near-perfect accuracy. Yet their decisions remain largely opaque at the signal level, so engineers cannot determine where in the raw vibration waveform the model focuses. This study aims to bridge the interpretability gap in deep learning-based bearing fault diagnosis by developing a signal-level explainability framework for the Wide Deep Convolutional Neural Network (WDCNN) on the Case Western Reserve University (CWRU) dataset. SHAP DeepExplainer is applied directly to raw 2,048-point vibration segments to produce per-sample-point attribution maps; Fault Signature Maps (FSMs) are formalized as class-averaged SHAP fingerprints in three variants (signed, absolute, and variance) and validated via discriminability index, severity monotonicity, and split-half stability, complemented by a three-variant ablation study examining how architectural decisions affect both accuracy and explainability. On the 10-class CWRU dataset under a rigorous temporal split (1,240/310/750 samples), WDCNN achieves 99.87% test accuracy with a macro F1-score of 0.997. FSMs demonstrate high reproducibility (split-half stability = 0.940); severity monotonicity ranges from 8.6% to 17.6%; and ablation reveals that removing Batch Normalization increases FSM discriminability by 2.3× at a 3.6% accuracy cost. As the first work applying SHAP DeepExplainer at full 2,048-point raw signal resolution for WDCNN, this study establishes that fault discrimination relies on transient impulse morphology rather than bearing characteristic frequencies, a finding invisible to feature-level XAI, and introduces a multi-resolution diagnostic framework bridging deep learning accuracy with physically interpretable vibration analysis for trustworthy deployment in safety-critical industrial environments.
- Open Access
High-Quality Foaming and Weight Reduction in Microcellular-Injection-Molded Polycarbonate Using Supercritical Fluid Carbon Dioxide under Gas Counter Pressure
Microcellular injection molding (MuCell®) using supercritical fluid (SCF) as a foaming agent to achieve weight reduction has become popular in carbon emission reduction. In the typical MuCell® process, SCF N2 is commonly used. Although SCF CO2 exhibits high solubility and can achieve a high weight reduction, controlling the foaming is not easy, and its foaming cells are usually larger and less uniform, which limits its industrial application. Our previous studies have shown that gas counter pressure (GCP) can improve the foaming quality effectively. Here, we investigated whether or not the CO2 SCF foaming quality could be improved, and weight reduction was achieved for polycarbonate (PC) material. This is quite important for the electronics industry, in which most of the housing for devices is made of PC materials. MuCell® was subjected to molding experiments using the parameters of the SCF dosage, melt temperature, mold temperature, and injection speed. The results revealed that using CO2 gas for the PC material can reduce the size of microcellular cells to 40 µm and increase the cell densities to 3.97 × 106 cells/cm3. Using GCP significantly improved the microcellular injection-molded parts by reducing the cell size to 20.9 µm (a 45.41% improvement) and increasing the cell density to 8.04 × 106 cells/cm3 (a 102.48% improvement). However, implementing GCP may slightly decrease the target weight reduction. This study reveals that microcellular injection molding of PC parts using SCF CO2 can achieve high-quality foaming and reduce the weight by about 30%.
Open AccessHigh-Quality Foaming and Weight Reduction in Microcellular-Injection-Molded Polycarbonate Using Supercritical Fluid Carbon Dioxide under Gas Counter Pressure
Microcellular injection molding (MuCell®) using supercritical fluid (SCF) as a foaming agent to achieve weight reduction has become popular in carbon emission reduction. In the typical MuCell® process, SCF N2 is commonly used. Although SCF CO2 exhibits high solubility and can achieve a high weight reduction, controlling the foaming is not easy, and its foaming cells are usually larger and less uniform, which limits its industrial application. Our previous studies have shown that gas counter pressure (GCP) can improve the foaming quality effectively. Here, we investigated whether or not the CO2 SCF foaming quality could be improved, and weight reduction was achieved for polycarbonate (PC) material. This is quite important for the electronics industry, in which most of the housing for devices is made of PC materials. MuCell® was subjected to molding experiments using the parameters of the SCF dosage, melt temperature, mold temperature, and injection speed. The results revealed that using CO2 gas for the PC material can reduce the size of microcellular cells to 40 µm and increase the cell densities to 3.97 × 106 cells/cm3. Using GCP significantly improved the microcellular injection-molded parts by reducing the cell size to 20.9 µm (a 45.41% improvement) and increasing the cell density to 8.04 × 106 cells/cm3 (a 102.48% improvement). However, implementing GCP may slightly decrease the target weight reduction. This study reveals that microcellular injection molding of PC parts using SCF CO2 can achieve high-quality foaming and reduce the weight by about 30%.
- Open Access
- 9 November 2023
- Published by Atlantis Press SARL
- p. 453-459
A Review on Real Driving Cycle-Based State of Charge Prediction for EV Batteries
- 9 November 2023
- Published by Atlantis Press SARL
- p. 453-459
Open AccessA Review on Real Driving Cycle-Based State of Charge Prediction for EV Batteries
- Open Access
- 10 May 2023
- Published by IEOM Society International
- Vol. 12 (6)
Predictive Maintenance in Bearing Production based on Machine Condition and Product Quality Data using Machine Learning Approach
- 10 May 2023
- Published by IEOM Society International
- Vol. 12 (6)
Open AccessPredictive Maintenance in Bearing Production based on Machine Condition and Product Quality Data using Machine Learning Approach
- 3 April 2023
- Published by International Journal of Technology in International Journal of Technology
- Vol. 14 (2) , 363-373
A Two-dimensional Maintenance Service Contract Considering Availability and Maintenance Cost
In this paper, we study a two- dimensional Maintenance Service Contract (MSC) characterized by two limits (dimensions) of age and usage. It is considered that an agent offers a two-dimensional MSC by guaranteeing a certain level of equipment available to consumers. The agent needs to reduce the total maintenance cost to offer competitive MSC prices. Preventive maintenance actions (PM) are periodically carried out, and each PM action is considered to improve reliability modeled by reducing the failure rate function. Two decision variables (the PM interval (T) and the reduction in the intensity function are obtained by considering two performance measures that are relevant to agents and consumers (i.e., availability and total maintenance cost). A numerical example is presented by considering three types of equipment usage rates: low, medium, and high. The optimization of the two performance measures can ensure availability targets and, at the same time, minimize total maintenance costs.
- 3 April 2023
- Published by International Journal of Technology in International Journal of Technology
- Vol. 14 (2) , 363-373
A Two-dimensional Maintenance Service Contract Considering Availability and Maintenance Cost
In this paper, we study a two- dimensional Maintenance Service Contract (MSC) characterized by two limits (dimensions) of age and usage. It is considered that an agent offers a two-dimensional MSC by guaranteeing a certain level of equipment available to consumers. The agent needs to reduce the total maintenance cost to offer competitive MSC prices. Preventive maintenance actions (PM) are periodically carried out, and each PM action is considered to improve reliability modeled by reducing the failure rate function. Two decision variables (the PM interval (T) and the reduction in the intensity function are obtained by considering two performance measures that are relevant to agents and consumers (i.e., availability and total maintenance cost). A numerical example is presented by considering three types of equipment usage rates: low, medium, and high. The optimization of the two performance measures can ensure availability targets and, at the same time, minimize total maintenance costs.
- 27 October 2019
- Published by Springer Nature
- p. 355-361
Nash Game Theory Leasing Contract Model of New and Recondition Complex Equipment
- 27 October 2019
- Published by Springer Nature
- p. 355-361
Nash Game Theory Leasing Contract Model of New and Recondition Complex Equipment
- 27 October 2019
- Published by Springer Nature
- p. 24-29
Optimal Down-Time Target for Performance Based Remanufactured Lease Contract
- 27 October 2019
- Published by Springer Nature
- p. 24-29
Optimal Down-Time Target for Performance Based Remanufactured Lease Contract
- 27 October 2019
- Published by Springer Nature
- p. 342-348
Dump Truck Maintenance Contract Model Considering Operational Conditions (Load, Road Inclination and Environment Condition)
- 27 October 2019
- Published by Springer Nature
- p. 342-348
Dump Truck Maintenance Contract Model Considering Operational Conditions (Load, Road Inclination and Environment Condition)
- Open Access
- 1 August 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1245 (1) , 012052
Optimal decision for two dimensional maintenance service contract involving three parties
A two-dimensional (i.e. time and use) maintenance service contract involving three parties i.e. manufacturer, agent and consumer are considered in this paper. The contract period by the manufacturer/agent begins since the warranty expires (W) until the lifetime of the equipment expires (L). The use of equipment by consumer is modeled with accelerated failure time (AFT) model. AFT model specifies the limit value of permitted equipment usage (Ul ). The contract time is limited to t = L, but equipment usage may exceed U l . High use of equipment will increase consumer revenue. However, the consumer will incur additional costs paid to the manufacturer/agent if the use of equipment exceeds U l . Therefore, consumers will seek the optimum value of use (U∗) that maximizes the profits. As with consumer, manufacturer and agent also want to maximize their profits. Decision problems in maintenance service contracts involving the three parties will be resolved consider the interaction of the manufacturer and the agent in carrying out maintenance action. One level optimization used if the manufacturer and agent compete with each other. However, if the manufacturer and agent interact in a cooperation, then optimization of the decision model becomes more difficult and should be done with bilevel optimization.
- 1 August 2019
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1245 (1) , 012052
Open AccessOptimal decision for two dimensional maintenance service contract involving three parties
A two-dimensional (i.e. time and use) maintenance service contract involving three parties i.e. manufacturer, agent and consumer are considered in this paper. The contract period by the manufacturer/agent begins since the warranty expires (W) until the lifetime of the equipment expires (L). The use of equipment by consumer is modeled with accelerated failure time (AFT) model. AFT model specifies the limit value of permitted equipment usage (Ul ). The contract time is limited to t = L, but equipment usage may exceed U l . High use of equipment will increase consumer revenue. However, the consumer will incur additional costs paid to the manufacturer/agent if the use of equipment exceeds U l . Therefore, consumers will seek the optimum value of use (U∗) that maximizes the profits. As with consumer, manufacturer and agent also want to maximize their profits. Decision problems in maintenance service contracts involving the three parties will be resolved consider the interaction of the manufacturer and the agent in carrying out maintenance action. One level optimization used if the manufacturer and agent compete with each other. However, if the manufacturer and agent interact in a cooperation, then optimization of the decision model becomes more difficult and should be done with bilevel optimization.
- 25 August 2018
- Published by Springer Nature
- p. 198-205
Lease Contract in Usage Based Remanufactured Equipment Service System
- 25 August 2018
- Published by Springer Nature
- p. 198-205
Lease Contract in Usage Based Remanufactured Equipment Service System