Effective Traffic Routing Under Disastrous Events Through Machine Learning

Author

• Eric Itemuagbor, Prairie View A&M University

Date of Award

8-2026

Document Type

Thesis

Degree Name

Master of Science

Department

Computer Science

Abstract

Disastrous events such as hurricanes, wildfires, and floods routinely disrupt transportation networks, preventing emergency response, evacuation, and the delivery of critical supplies. Conventional optimization-based routing methods offer strong theoretical guarantees but cannot scale and are often too slow to support time-sensitive decision-making during a disaster, while prior machine learning approaches to disaster routing have largely prioritized prediction accuracy over inference speed and data efficiency. This thesis investigated whether lightweight machine learning models can deliver both competitive routing accuracy and superior inference speed when trained on limited data, addressing a practical gap between existing research and real-world deployment needs.

The road network was modeled as a synthetic abstraction of the Houston metropolitan highway system, comprising 22 nodes and 38 bidirectional edges, and the routing problem was formulated as a Minimum-Cost Flow problem. Disaster impact was simulated by systematically perturbing edge costs across twelve road groups, yielding 531,441 distinct disaster scenarios used to generate training and testing data. Lightweight machine learning models such as Random Forest with Regressor Chain, Bidirectional Long Short-Term Memory (BiLSTM), Transformer, and a policy-gradient Reinforcement Learning model were benchmarked against classical optimization and approximation algorithms such as Linear Programming Relaxation, Successive Shortest Path, and Cost-Scaling Push Relabel.

Results showed that the BiLSTM model achieved the highest routing accuracy among the learning-based methods, reaching 75.26% correct routing with only 30% of available training data, while the Reinforcement Learning model achieved the fastest inference time at a moderate accuracy trade-off. All machine learning models substantially outperformed classical algorithms in inference speed, though the latter retain guaranteed optimality. These findings demonstrated that selective lightweight machine learning models can serve as practical, time-efficient alternatives for disaster-impacted traffic routing where rapid decision-making is essential.

Index Terms—Bidirectional long short-term memory (BiLSTM), data efficiency, disaster response, machine learning, minimum-cost flow, random forest, real-time decision making, regressor chain, reinforcement learning, traffic routing, transformer

Committee Chair/Advisor

Lin Li

Committee Co-Chair

Lijun Qian

Committee Member

Sherri S. Frizell

Committee Member

Md Hossain Shuvo

Publisher

Prairie View A&M University

Rights

Date of Digitization

8/27/2026

Contributing Institution

J. B Coleman Library

City of Publication

Prairie View

MIME Type

Application/PDF

Recommended Citation

Itemuagbor, E. (2026). Effective Traffic Routing Under Disastrous Events Through Machine Learning. Retrieved from https://digitalcommons.pvamu.edu/pvamu-theses/1684