The role of statistics in operations research: Some personal re ections

Abstract

Statistics has a very important role to play in Operations Research (OR), yet many standard texts relegate it to a subordinate position behind, for example, deterministic optimisation techniques. I argue in this paper that statistics, and statistical thinking, are essential for the OR practitioner operating in the real, uncertain world. Using many examples from my own
experience, I demonstrate that statistics is very useful, and indeed essential, when tackling real-world OR problems. Many of these examples have not been published before and will therefore hopefully be of interest in themselves, as well as illustrations of the points I wish to make.

Key words: Statistical thinking, problem formulation, Bayesian analysis.