Decision-Making Support in Complex Multi-Actor and Multi-Source Scenarios
Álvarez Márquez, J. O. (2022). [PhD thesis].
Abstract
The term “complex decision environment” has been used in literature to describe situations that involve making a decision with one or more of the following aspects: multiple stakeholders, multiple viable alternatives, contradictory or incomplete information, multiple sources, and dynamic settings. This thesis aims at achieving a better understanding of the processes that take place in or under complex decision environments, with a focus on cases where multiple actors or multiple sources of information are present. To that end, the first goal of this research work is to define specific but sufficiently common scenarios in which multi-factor decision-making takes place. Examining these scenarios can lead to a greater comprehension of the particular requirements, decision processes and challenges faced by decision-makers when caught in a multi-factor decision environment. After defining these scenarios and their characteristics, a second research goal is to design and evaluate suitable methods to support the decision-making process that takes place in them. The thesis presents investigations and developments in two different application contexts. The first one concentrates on complex multi-actor decision environments, illustrated by a group of non-collocated people who must jointly choose a hotel to stay together. The second scenario deals with complex multi-source decision environments, and more specifically with consumers facing a purchase decision in a physical store setting where online information is also available. Methods have been designed to assist decision-makers by taking into account the unique characteristics of each scenario, for which the use of recommender systems and technological innovations (especially in relation to augmented reality) has proven to be advantageous. The evaluation of the developed methods provides insight into the cognitive processes and constraints that emerge in multifactorial settings, and establishes design guidelines for future research.