Ata uhi

Advanced Sales Route Optimization Through Enhanced Genetic Algorithms and Real-Time Navigation Systems

I tiakina i:
I whakaputaina i:Algorithms vol. 18, no. 5 (2025), p. 260
Kaituhi matua: Cunuhay Cuchipe Wilmer Clemente
Ētahi atu kaituhi: Zajia, Johnny Bajaña, Oviedo Byron, Zambrano-Vega, Cristian
I whakaputaina:
MDPI AG
Ngā marau:
Navigation systems
Deep learning
Adaptability
Mathematical models
Traffic
Mutation
Travel time
Environmental impact
Energy consumption
Sales
Machine learning
Dynamic programming
Genetic algorithms
Route optimization
Prediction models
Route planning
Tabu search
Neural networks
Optimization
Travel
Linear programming
Customers
Real time
Logistics
Traveling salesman problem
Urunga tuihono:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Ngā Tūtohu: Tāpirihia he Tūtohu
Kāore He Tūtohu, Me noho koe te mea tuatahi ki te tūtohu i tēnei pūkete!
Whakarāpopotonga:Efficient sales route optimization is a critical challenge in logistics and distribution, especially under real-world conditions involving traffic variability and dynamic constraints. This study proposes a novel Hybrid Genetic Algorithm (GAAM-TS) that integrates Adaptive Mutation, Tabu Search, and an LSTM-based travel time prediction model to enable real-time, intelligent route planning. The approach addresses the limitations of traditional genetic algorithms by enhancing solution quality, maintaining population diversity, and incorporating data-driven traffic estimations via deep learning. Experimental results on real-world data from the NYC Taxi dataset show that GAAM-TS significantly outperforms both Standard GA and GA-AM variants, achieving up to 20% improvement in travel efficiency while maintaining robustness across problem sizes. Although GAAM-TS incurs higher computational costs, it is best suited for offline or batch optimization scenarios, whereas GA-AM provides a balanced alternative for near-real-time applications. The proposed methodology is applicable to last-mile delivery, fleet routing, and sales territory management, offering a scalable and adaptive solution. Future work will explore parallelization strategies and multi-objective extensions for sustainability-aware routing.
ISSN:1999-4893
DOI:10.3390/a18050260
Puna:Engineering Database