Title: Optimal purchasing strategy for bandwidth: a buyer's perspective

Authors: Elizabeth Sharer; Hari K. Rajagopalan; Sanjaya Mayadunne

Addresses: School of Business, Francis Marion University, Florence SC, 29506, USA ' School of Business, Francis Marion University, Florence SC, 29506, USA ' School of Business, Georgia Gwinnett College, Lawrenceville, GA 30043, USA

Abstract: The price of internet bandwidth contracts depends on two parameters: size, or amount of bandwidth, and duration, or length of contracts. For a given planning horizon, managers must decide on the number of contracts to purchase, their bandwidth (size), and their lengths (duration) to minimise costs. This research analyses bandwidth contracting decisions from a buyer's perspective. Historical data is used to estimate bandwidth cost as a function of contract size and duration at a point in time, and to estimate bandwidth cost overtime. We first build an integer linear programming model, and then show that evolutionary algorithms can match the performance of mathematical programming and achieve close to optimal solutions for problems under increasing bandwidth demand.

Keywords: bandwidth contract; mixed integer linear programming; evolutionary algorithms.

DOI: 10.1504/IJPMB.2025.143985

International Journal of Process Management and Benchmarking, 2025 Vol.19 No.2, pp.238 - 265

Received: 23 May 2023
Accepted: 16 Jun 2023
Published online: 17 Jan 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article