Thursday, December 5 2019

12:30pm - 1:00pm

12:30pm - 1:00pm

Operations Research Seminar

Using Total Unimodularity to Solve a Warehouse Problem

With a daily increase in the usage of online retailers like Amazon, the operations of warehouses

are becoming more and more valuable. One way to create an efficient warehouse is to effectively place items that have yet to be shipped. The primary goal of this project is to develop a mathematical model that could create a scheme of inventory layout in a warehouse.

From a mathematical standpoint, the original model we propose is an integer programming model.

In order to solve any integer program of a reasonable size, one would need a significant amount of memory and a significant amount of time. A natural next step then is to relax the developed integer program for a more efficient solution. We will do so by using a mathematical concept called “totally unimodularity” to prove that the optimal vertex solutions of a linear programming relaxation are integral.

are becoming more and more valuable. One way to create an efficient warehouse is to effectively place items that have yet to be shipped. The primary goal of this project is to develop a mathematical model that could create a scheme of inventory layout in a warehouse.

From a mathematical standpoint, the original model we propose is an integer programming model.

In order to solve any integer program of a reasonable size, one would need a significant amount of memory and a significant amount of time. A natural next step then is to relax the developed integer program for a more efficient solution. We will do so by using a mathematical concept called “totally unimodularity” to prove that the optimal vertex solutions of a linear programming relaxation are integral.

Speaker: | Zachary Sorenson |

Affiliation: | University of Colorado Denver |

Location: | SCB 4119 |

Done