Abstract
Operations automation is a concept and design methodology for human-centered automation. Operations automation is automation that carries out, in whole or part, activities currently performed by operations personnel. Humans, however, remain essential components of the system, but perform a significantly different role. When autonomous systems reach their inevitable limits, operations personnel troubleshoot, diagnose, and repair the automation.
This paper describes research exploring requirements for effective operations automation. First, it describes two field studies of conventional automation. Second, it proposes operations automation as an extension of the operator function model (OFM) and its computational implementation, OFMspert. The OFM is a normative behavioral model; OFMspert predicts and interprets operation actions. As such, the OFM and OFMspert offer potential computational architectures with which to implement operations automation. Finally, the paper describes an empirical study comparing conventional and operations automation. Results suggest operations automation dramatically enhances the ability of operations personnel to identify and diagnose failures.
Get full access to this article
View all access options for this article.