Engineering and Technology Horizons

A Novel Weighted Ebola Vector Optimization Based Charging Scheduling of Electric Vehicle Integrated with PV Charging Stations

Nageswara Rao — 2026-07-27

Electric vehicle (EV) charging scheduling integrated with photovoltaic (PV)-based charging stations is an important aspect of smart energy management. This paper presents an optimal EV charging scheduling framework considering solar-powered charging infrastructure and bidirectional vehicle-to-grid (V2G) and grid-to-vehicle (G2V) power transfer. A hybrid weighted ebola vector algorithm is proposed to improve charging coordination and minimize operational cost. In addition, photovoltaic power generation is forecasted using an artificial neural network (ANN) for accurate 24-hour solar energy prediction. The objective function considers peak shaving, valley filling, power loss, charging coordination, and charging cost minimization under varying load conditions. The proposed method is evaluated using a 24-hour load duration curve with forecasted PV power and power loss is compared with the Ebola Optimization Search Algorithm and Weighted Mean Vector Optimization (WMVO) technique under Normal, High EV plug-in, Low Solar power availability, and Peak Hour scenarios. Simulation results demonstrate that the proposed algorithm achieves superior performance in convergence speed, scheduling accuracy, and power loss reduction. The proposed method reduces charging cost by 28%, whereas the Ebola and WMVO methods achieve reductions of 17.7% and 8.8%, respectively. The results confirm the effectiveness of the proposed optimization approach for smart EV charging management with renewable energy integration.

Comparative Study on the Enhancement of Soil Cement Strength Using Bentonite and Polyurethane Foam

Pornkanok Rattanapituk — 2026-09-03

The construction of buildings and infrastructure on soft ground, such as soft Bangkok clay, presents significant geotechnical challenges due to its high compressibility, high settlement potential, and low bearing capacity. Consequently, soil-cement stabilization is widely applied to improve the strength of soft ground. This study evaluated the effects of bentonite and polyurethane foam as additive admixtures to increase the strength of soil cement. The primary purpose was to investigate compressive strength. Moreover, Scanning Electron Microscopy (SEM) was employed to examine the microstructure, which supports the primary purpose. The control specimen consisted of 80% soil and 20% cement by dry weight of the total soil-cement mixture. Bentonite and polyurethane foam contents of 10%, 20%, and 30% by dry weight of the total soil-cement mixture were investigated and added as additive admixtures to the soil-cement mixture. The results showed that adding bentonite and polyurethane foam can increase the strength of soil cement. The suitable ratios of bentonite and polyurethane foam were 30% and 10% by dry weight of the total soil-cement mixture, respectively. Moreover, both bentonite and polyurethane foam can increase compressive strength. This approach offers a pathway to more efficient, sustainable soil-cement stabilization methods to improve soft soils.

An Investigation of the Relationships between Road Geometric Characteristics, Vehicle Speed, Road Lighting Conditions and Traffic Crash Occurrence for Road Safety Design

Sirinapa Jantarakot — 2026-09-03

This study aims to: 1) examine the relationship between the physical characteristics of roads, road lighting conditions, vehicle speed, and road crash data; and 2) propose guidelines for geometric road design to improve road safety. The study area focuses on Sri Chan Road in Khon Kaen Municipality, Khon Kaen Province, Thailand. Field data collection included the physical characteristics of the road, vehicle speed data, road lighting conditions, and road crash data. The data was analyzed using descriptive statistics, the Accident Frequency Method, and Pearson Product Moment Correlation and Simple Linear Regression Models to explore relationships between variables. Geometric designs were also developed according to engineering principles. The study found that the mean speed of private cars was 35.73 km/h per section of the road (SD. = 4.48), and the mean speed of motorcycles was 35.24 km/h per section (SD. = 3.88). The area studied had an average number of injuries of 41.91 people/road sections. Road width was significantly related to the number of lanes, the presence of a median, and the average speed of private cars and motorcycles. The 85th percentile speed of private cars and the 85th percentile speed of motorcycles was also statistically correlated (p-value < 0.05). Additionally, the intensity of road lighting was negatively correlated with sections where severe road crashes occurred, resulting in fatalities (r = -0.702, p-value < 0.05). Similarly, the results of the Simple Linear Regression Models indicate that the mean vehicle speed is significantly related to the number of traffic lanes at a statistically significant level. Based on these results, geometric designs were proposed to reduce vehicle speeds, such as speed-reducing lines, speed limits near important places, and roundabouts. The findings will be submitted to relevant local authorities for future road safety improvements.

Evaluating the Effectiveness of Artificial Intelligence (AI) Cameras in Detecting Helmet Use

Jetsada Kumphong — 2026-09-18

Currently, Thailand has a higher death rate from motorcycle accidents than from other vehicles. Wearing a helmet can reduce the severity of head injuries in accidents, since the head is a crucial part of the human body. The objective of this research was to evaluate the effectiveness of AI camera detection of helmet-wearing behaviours (both drivers and passengers) and to find the relationship between the accuracy of AI cameras in detecting helmet-wearing behaviours and the camera installation points. Data was collected from two AI camera installation points, observing a total of 674 motorcycles. The data was analyzed using descriptive statistics and the Pearson Chi-Square test, as well as a logistic regression analysis. The study found that AI cameras could accurately detect helmet-wearing behaviours, with Point 1 having a detection accuracy of 73.7% and Point 2 having a detection accuracy of 89.5% (Point 1 was a two-lane, one-way roadway featuring a covered pedestrian walkway with high foot traffic; whereas Point 2 was a two-lane, bi-directional roadway flanked by drainage ditches on both sides with no dedicated pedestrian facilities.). Additionally, there was a significant difference in the detection accuracy of the AI cameras between the two installation points (p-value < 0.05), as well as a difference in the detection accuracy based on helmet-wearing behaviours (p-value < 0.05). In addition, analysis revealed that camera location and helmet type had significantly impacted detection accuracy. Specifically, Point 2 outperformed Point 1, and Type 2 helmet detection showed a 23-fold higher accuracy compared to other types (p-value < 0.05)

Machine Learning–Driven Predictive Maintenance for Smart Beverage Manufacturing Systems

Peerawit Pongtananikorn — 2026-09-03

This study proposed a machine learning-based predictive maintenance system in a beverage manufacturing process to enhance failure prediction accuracy and support improved maintenance planning. The approach focused on the beverage filling line, captured short-term temporal process dynamics using rolling-window feature extraction, and addressed the strong class imbalance between normal operation and breakdown events. Significant sensor variables were initially identified through correlation analysis, and the selected sensors and breakdown criteria were subsequently analyzed using various classification models. A 60-minute decision horizon was defined as a classification-based early-warning window derived from the time-to-failure (TTF) variable. Observations were labeled breakdown-proximate state when the next breakdown occurred within 60 minutes (TTF < 60 min) and as normal operation when the next breakdown occurred after 60 minutes (TTF > 60 min). Therefore, the model classifies a breakdown-proximate state rather than predicting the exact remaining time to failure. Model performance was evaluated using precision, recall, F1-score, ROC-AUC, and chronological walk-forward validation. Based on correlation analysis and process relevance, 15 sensors were selected as key features associated with process breakdown. The Random Forest model was the best performer, achieving an accuracy of 0.990, precision of 0.918, recall of 0.722, F1-score of 0.808, and ROC-AUC of 0.995 at the optimized threshold of 0.45. The recall result indicates acceptable detection of breakdown-proximate states, although 27.8% was missed as a false negative. Considering the balance between recall and precision, the precision exceeded 90%, showing that false alarms were maintained at a low level. These results demonstrate a practical trade-off between reducing missed breakdown warnings and avoiding excessive unnecessary inspections. Thus, the Random Forest is suitable for predicting breakdowns with accuracy, stability and robustness. Future work will extend the predictive maintenance results linked with failure-risk predictions to reliability metrics of time between failures (TBF) and mean time between failures (MTBF), improved alignment between maintenance actions and actual machine conditions.

Semi-Automated Air-Cell Image Measurement and Weight-Loss Monitoring for Non-Destructive Assessment of Egg Freshness During Storage

Mookarin Nookong — 2026-09-03

Conventional methods for evaluating egg freshness rely heavily on candling, visual inspection, or destructive laboratory tests. These approaches are often subjective, operator-dependent, and difficult to apply consistently in large-scale food production. To address these limitations, this study developed a non-destructive approach for monitoring egg freshness-related changes during storage by combining cumulative weight-loss measurement with image-based air-cell area measurement. A total of 100 eggs were stored under ambient conditions and repeatedly measured at five time points: Day 1, Day 6, Day 11, Day 16, and Day 21. Repeated-measures ANOVA confirmed significant effects of storage duration on egg weight and air-cell area (p < 0.001). Cumulative weight loss reached 4.22% by Day 21, while the mean air-cell area increased from 90.6 mm² on Day 1 to 274.7 mm² on Day 21 after skipped air-cell annotations were treated as missing data. Cumulative weight loss showed a significant positive association with air-cell area ratio when individual egg-level observations were analyzed. However, the moderate regression strength indicated that air-cell enlargement was not explained by weight loss alone and may also be influenced by egg-specific factors such as shell structure, shell porosity, and initial egg size. These findings indicate that combining weight-loss monitoring with GUI-based air-cell image measurement can support non-destructive assessment of freshness-related changes in eggs during storage.

Degradation-Aware Single-Machine Scheduling with Price-Driven Thresholding and Waiting Triggers

Zhaoyi Zhang — 2026-09-18

This paper studies a degradation-aware non-preemptive single-machine scheduling problem under time-of-use (TOU) and flat (FLAT) electricity tariffs. The objective is to minimize total cost, including processing energy, optional waiting energy, preventive maintenance (PM), and weighted tardiness. The model links reliability degradation to processing power through an interpretable reliability-power mapping and uses an MTBF-calibrated exponential reliability function with PM restoration. We propose a price-driven adaptive reliability threshold rₜₕ(t), bounded by rₘᵢₙ and rₘₐₓ, and a conservative waiting trigger that is activated only when due-date slack is sufficient and the expected price decrease is meaningful. Under FLAT pricing, the adaptive term vanishes and the policy degenerates to the static baseline, providing a direct consistency check. For computation, PM and waiting decisions are generated by policy rules, whereas metaheuristics optimize the job permutation. A permutation-based Tabu Search (TS) is compared with GA, SA, ACO, and DE-RK on OR-Library WT40/WT50/WT100 instances. The results show rational TOU responsiveness, small-instance optimality references, and non-additive interactions between threshold adaptiveness and waiting under strong due-date pressure. The framework provides auditable decision logic for scheduling and maintenance under time-varying electricity prices.

Integer Linear Programming Models for Preventive Maintenance of Medical Devices: A Case Study

Todsapol Huana — 2026-09-18

This study aims to develop a mathematical decision-support framework for optimizing workforce allocation and operational planning in medical device preventive maintenance (PM). An investigation a case study at a public hospital revealed that the PM plan for medical devices at the operational level often lacks decision-support tools for effective work planning. Due to the large number and diverse range of medical devices in hospitals, the head of the medical device department often encounters difficulties in allocating technicians efficiently across different device categories according to their skill levels and time constraints. This limitation frequently results in disproportionate workloads and an inequitable distribution of tasks among personnel. To address these inefficiencies, this study employs a sequential optimization framework consisting of a Binary Integer Linear Programming (BILP) model followed by an Integer Linear Programming (ILP) model. First, the BILP model assigns maintenance tasks based on technician skills and time constraints with the simultaneous optimization of workload balance and equitable distribution of maintenance tasks among technicians within a single model. The optimized BILP output is then used as input for the ILP model to determine optimal daily task allocation. This sequential approach effectively reduces computational complexity. The proposed models are implemented within Microsoft Excel and integrated with OpenSolver, a free and open-source optimization solver, to provide a flexible and cost-effective alternative to manual planning. The results indicate that the proposed BILP and ILP models reduced labor costs by 12.64% and 12.60%, respectively, compared with the current maintenance planning. The proposed model effectively balances workloads among technicians within the same skill group, resulting in workload deviations of 0.70% and 0.84% for the skilled and general groups, respectively. Furthermore, equitable task distribution is achieved, as the difference in the number of assigned devices of each type among technicians within the same skill group does not exceed one device.

Polarization Reconfigurable Dipole Antenna for FM Band Radio Spectrum Monitoring System

Nutdanai Hantankul — 2026-09-18

This paper presents the design and implementation of a polarization reconfigurable dipole antenna operating in the FM radio frequency band (88–108 MHz) for low-cost radio spectrum monitoring applications. The proposed antenna consists of two orthogonal radiating arms integrated with an electronically controlled switching network that enables selection between two linear polarization states. Simulated results show that the antenna achieves |S₁₁| below −10 dB across the entire FM band, with stable omnidirectional radiation patterns and realized gains of approximately 1.7 dBi and 1.8 dBi for the two polarization modes. A prototype was fabricated and measured to validate the design. Measured results confirm acceptable impedance matching over the frequency range of 93–102 MHz, covering the core FM broadcast channels, while maintaining omnidirectional radiation characteristics and effective polarization switching. The minor frequency shift observed in measurements is attributed to fabrication tolerances and environmental effects. Owing to its simple structure, low fabrication cost, and reliable performance, the proposed antenna is well-suited for large-scale deployment in automated FM-band radio spectrum monitoring systems.