Minimum Variance Formulation for the Stochastic Vehicle Routing Problem with Uncertain Demands

Project start: 2018, Project end: 2018
Funding agency: Erasmus+

Project description

The Vehicle Routing Problem (VRP) is a combinatorial optimization problem which seeks to design a routing plan to attend a set of geographically dispersed customers with a given vehicle fleet at minimum cost. The Stochastic VRP happens when one or more inputs present uncertainties, i.e. they are known as stochastic variables.
This project is in the area of stochastic capacitated vehicle routing problem. We use robust optimization to calculate a solution which is robust against some changes in the customer's demand. Instead of using the most common formulation min-max we propose a mean-variance formulation. The goal is to allow decision makers to choose the degree of robustness for the solutions calculated for the problem.
Contact: Nazlihan Mustu