Developing Optimization Models with Cognitive Systems Engineering

TL;DR

This paper proposes a framework that integrates Cognitive Systems Engineering methods into optimization model development to enhance decision-making and stakeholder understanding, demonstrated through a COVID-19 sanitizer station case study.

Contribution

It introduces a novel framework combining cognitive engineering and optimization modeling to improve model relevance and stakeholder engagement in applied operations research.

Findings

Enhanced stakeholder understanding of models

Improved alignment between models and system dynamics

Successful application in COVID-19 sanitizer station placement

Abstract

One goal of applied operations research is to improve decisions in practice. This requires modelers and stakeholders to have a shared understanding of the system and for the developed model to reflect the system's core dynamics. There are four areas to address: the underlying problem must be understood, the mathematical formulation of the problem must be representative of the system at hand, the data must be appropriate, and the model-generated recommendations must be understandable by the stakeholders. While developing models, operations researchers may primarily rely on past experience in model development, rather than underlying theory, to guide decisions on how to include stakeholders in the modeling process. In parallel, the field of Cognitive Systems Engineering has developed methodologies and practices to understand systems, stakeholder needs, and environments. To improve the rigor of the "application" in applied operations research, we present a framework to integrate Cognitive Systems Engineering methods with optimization model development. We apply the integrated framework to a case study of locating hand sanitizer stations in response to COVID-19 at a large academic institution.