Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Journal of Applied Data Sciences

The Cumulative Capacitated Vehicle Routing Problem with Time-dependent on Humanitarian Logistics for Disaster Management Hartama, Dedy; Wanayumini, Wanayumini; Damanik, Irfan Sudahri
Journal of Applied Data Sciences Vol 6, No 1: JANUARY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i1.481

Abstract

This study addresses the challenges of optimizing humanitarian logistics during disaster management by developing a Cumulative Capacitated Vehicle Routing Problem with Time-Dependent factors (CCVRP-TD) model. The primary objective is to enhance delivery efficiency by incorporating time-dependent variables such as fluctuating traffic and service durations into route planning. The research contributes a novel Mixed Integer Nonlinear Programming (MINLP) framework that dynamically adapts to real-world conditions like road closures and shifting priorities. Using the MINLP approach, the model was validated through numerical experiments involving four delivery vehicles serving six customers across five routes. Results demonstrated a significant improvement in routing efficiency, with a total cumulative travel distance of 110 km and adherence to specified delivery windows, such as 9:30 AM and 10:30 AM for Customer 1. Additionally, vehicle capacity constraints were effectively managed, with individual route lengths ranging from 20 to 35 km. These findings showcase the model’s ability to balance cost minimization, service reliability, and logistical adaptability. The novelty lies in the integration of time-dependent costs and service benefits into a multi-depot framework, enabling flexible yet precise route optimization under constrained conditions. This research provides a robust tool for enhancing disaster logistics and offers practical implications for improving the responsiveness and effectiveness of humanitarian aid delivery.
A Hybrid GRG-Neighborhood Search Model for Dynamic Multi-Depot Vehicle Routing in Disaster Logistics Hartama, Dedy; Poningsih, Poningsih; Tanti, Lili
Journal of Applied Data Sciences Vol 6, No 4: December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i4.973

Abstract

In disaster relief logistics, timely and adaptive routing is critical to meet fluctuating demands and disrupted infrastructure. This paper proposes a Hybrid GRG–Neighbourhood Search (NS) model for solving the Multi-Depot Vehicle Routing Problem with Capacity and Time Dependency (MDVRP-CTD). The model integrates the Generalized Reduced Gradient (GRG) method for handling nonlinear capacity constraints and NS for local route refinement. The objective is to minimize total travel distance, delay penalties, and maximize vehicle utilization under dynamic disaster scenarios. Tested using the SVRPBench dataset, the hybrid model achieved up to 96.5% demand fulfillment, an 11% improvement in vehicle utilization, and a reduction in total distance by 7%, outperforming Tabu Search and ALNS in three simulation scenarios. The model demonstrates enhanced adaptability and responsiveness to time-sensitive, capacity-constrained environments. Its novelty lies in the integration of nonlinear optimization with adaptive local improvement tailored for disaster contexts, providing a robust decision-support tool for real-time humanitarian logistics.
Multiclass Skin Lesion Classification Algorithm using Attention-Based Vision Transformer with Metadata Fusion Furqan, Mhd.; Katuk, Norliza; Hartama, Dedy
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.1017

Abstract

Early and accurate classification of skin lesions is essential for timely diagnosis and treatment of skin cancer. This study presents a novel multiclass classification framework that integrates dermoscopic images with clinical metadata using an attention-based Vision Transformer (ViT) architecture. The proposed model incorporates a mutual-attention fusion mechanism to jointly learn from visual and tabular inputs, augmented by a class-aware metadata encoder and imbalance-sensitive loss function. Training was conducted using the HAM10000 dataset over 30 epochs with a batch size of 32, utilizing the Adam optimizer and a learning rate of 0.0001. The model demonstrated superior performance compared to a ViT Baseline, achieving 93.4% accuracy, 92.2% F1-score, 0.95 AUC, and significant reductions in MAE and RMSE. Additionally, Grad-CAM visualizations confirmed the model’s ability to focus on diagnostically relevant regions, enhancing interpretability. These findings suggest that the integration of structured clinical information with transformer-based visual analysis can significantly improve classification robustness, particularly in underrepresented lesion types. However, the model’s current performance is evaluated only on the HAM10000 dataset, and its generalizability to other clinical or non-dermoscopic image sources remains to be validated. Future studies should therefore explore multi-institutional datasets and real-world deployment scenarios to assess robustness and scalability. The proposed framework offers a practical, interpretable solution for AI-assisted skin lesion diagnosis and demonstrates strong potential for clinical deployment.