Machine learning and optimization for intelligent decision-making

Author

Elson Cibaku, New Jersey Institute of TechnologyFollow

Author ORCID Identifier

0000-0003-3161-3662

Document Type

Dissertation

Date of Award

5-31-2025

Degree Name

Doctor of Philosophy in Industrial Engineering - (Ph.D.)

Department

Mechanical and Industrial Engineering

First Advisor

SangWoo Park

Second Advisor

Esra Buyuktahtakin Toy

Third Advisor

Sanchoy K. Das

Fourth Advisor

Golgen Bengu

Fifth Advisor

Petras Juozas Swissler

Sixth Advisor

Marcos Netto

Abstract

This dissertation presents a series of innovative machine learning and optimization model designs that address complex operational challenges across logistics and power systems. By integrating advanced neural architectures with robust optimization techniques, the work delivers scalable solutions designed to improve efficiency, reliability, and decision-making in dynamic and real-world environments. The first study introduces a two-stage approach to effective vaccine distribution. This framework tackles the capacitated vehicle routing problem by combining adaptive clustering techniques with reinforcement learning and a simulated annealing pickup policy. Through extensive computational experiments, the approach demonstrates substantial improvements in routing efficiency, reducing both computational time and logistical costs while ensuring timely and reliable delivery under strict constraints in NP-hard settings. In the second study, a capacity-constrained K-means clustering method is proposed to optimize the multi-pickup and multi-delivery of restaurant orders. By intelligently clustering orders based on spatial and temporal criteria, the model efficiently organizes delivery routes to minimize delays and operational expenses. The simulation results underscore its effectiveness in improving resource utilization and responsiveness to service, promising an improved food service landscape with automated and data-driven route planning. The third study focuses on improving the efficiency of power system state estimation by leveraging the inherent graph structure of power grids combined with the attention mechanism. Here, a graph attention estimation network model is designed, which integrates graph convolutional neural networks with an attention-based feature selection mechanism.

Recommended Citation

Cibaku, Elson, "Machine learning and optimization for intelligent decision-making" (2025). Dissertations. 1831.
https://digitalcommons.njit.edu/dissertations/1831