Date of Award
2015
Document Type
Thesis
Degree Name
Bachelors
Department
Natural Sciences
First Advisor
McDonald, Patrick
Keywords
SBRP, School Bus Routing Program, Students, Algorithms
Area of Concentration
Applied Mathematics
Abstract
Given a collection of bus stops, each occupied by a number of students, and a fleet of school buses subject to operational constraints, the School Bus Routing Problem (SBRP) asks for the routing of buses which minimize a given objective function. It is considered to be np-hard and is a problem of enormous importance in the field of combinatorial optimization and has numerous applications in trans- portation logistics. We survey the literature relative to the SBRP and develop the machinery of linear programming in order to assess the validity of the algorithms used to solve our SBRP model. We introduce student risk as an objective and study the behavior of optimal routes as the relative cost of risk is varied. A commercial optimizer is used to solve our model and visualizations are created using the graphics package of the programming language R, which provide empirical evidence for our conjecture involving patterns in the change of route topology as a function of relative risk.
Recommended Citation
Kao, Roberto, "A numeric approach to analyzing the properties of solution changes in a multi-objective School Bus Routing Problem" (2015). Theses & ETDs. 5043.
https://digitalcommons.ncf.edu/theses_etds/5043