Research & Projects

Project 1

Post Disaster Routing Optimization (ongoing)

Abstract

Vehicle routing problems (VRPs) are critical in post-disaster scenarios, where efficient and effective deliveryof relief materials is paramount. This research introduces a novel approach to VRPs in post-disaster road networks, incorporating stochasticity at both link and node levels. The study proposes the use of ensemble learning to develop a new solving method that combines various effective algorithms for such problems. The goal is to produce optimal solutions efficiently, thereby improving the speed and effectiveness of disaster response. The research will explore the performance of this new method and its potential implications for disaster management and logistics planning.

Results

As it is a work in progress, specific results and findings are not yet available. Please stay tuned for updates as our work progresses, and I look forward to sharing our findings with you in the near future.

Simulation_Video.mp4

Project 2

Stochastic Optimization of Drone Delivery Nework (Ongoing)

Abstract

In this research, we explore a practical and forward-thinking solution to make drone deliveries more efficient on the Australian Post Dataset. Using the SIMIO simulation tool, we've modeled a scenario where drones operate with the support of a single charging station, tackling the real-world challenge of maintaining consistent service over vast distances. The heart of our project is the innovative use of Reinforcement Learning, a technique that helps drones learn and improve their delivery routes and timings. Our approach not only addresses the logistical hurdles like range limitations and energy management of drones but also dynamically adapts to changing postal service demands, paving the way for smarter and more effective postal delivery systems.

Results

Work in Progress.

Review Paper

TRANSPORTATION TRANSFORMED: A COMPREHENSIVE REVIEW OF DYNAMIC REROUTING IN MULTIMODAL NETWORKS

Abstract

The emergence of dynamic rerouting in multi-modal transportation networks has emerged as a crucial area in operations research, revolutionizing routine optimization. The review study analyzes 25–30 research publications on algorithms and techniques related to dynamic rerouting to give a thorough summary of the state of research in this field and provide future suggestions. The research paper explains the importance of dynamic rerouting in modern transportation systems and recognizes its critical role in tackling issues like accidents, traffic congestion, and infrastructure constraints. In addition, the review examines the development of dynamic rerouting techniques by examining several studies to uncover the theoretical foundation, technological developments, and effects of the practices on various forms of transportation. The paper emphasizes the potential of technological advancements such as artificial intelligence, the Internet of Things, and big data in transforming routing efficiencies. Further, the review presents specific difficulties and best practices for each mode of transportation, highlighting the many uses of dynamic rerouting in air, sea, rail, and road transportation. The review also digs deeper into the integration barriers common in multi-modal networks, highlighting successful case studies that overcome these obstacles as well as strategic approaches and regulatory modifications. Lastly, the research paper assesses the impact of dynamic rerouting on urban development, sustainability, and potential directions for future research such as the integration of large language models. The comprehensive literature review incorporates multiple research perspectives to offer significant insights into the efficacy, challenges, and potential future pathways for dynamic rerouting within multi-modal transportation networks.

Results

Pre-print : arxiv.org/abs/2312.14953

2312.14953.pdf

Page updated

Google Sites

Report abuse