How can they do it? A structured approach to modelling performance variability

2017-02-28T04:55:44Z (GMT) by Cornelissen, Miranda
Performance variability is a contemporary safety related concept in Ergonomics. A literature review revealed that performance variability is not well understood, particularly in road transport. Moreover, performance variability has mainly been studied in a negative context and using reductionist approaches. A methods review revealed that, outside of the Cognitive Work Analysis (CWA) framework, currently there is no structured method available to satisfactorily model performance variability in complex systems. To address these gaps, the research presented in this thesis aimed to develop, apply and evaluate a method for modelling performance variability. An initial application of CWA to describe performance variability in road transport, confirmed that performance variability is a natural occurrence in road transport and that theories of performance variability, such as the field of safe travel, fail to represent important aspects of performance variability. Importantly, the application revealed gaps in the CWA phase that is essential for modelling possible courses of action, the Strategies Analysis (StrA) phase. The Strategies Analysis Diagram (SAD) was developed to provide StrA and CWA with a structured method for modelling performance variability. An initial evaluation of SAD using a simple system, the Apple iPod, confirmed that SAD addressed the gaps identified in the StrA, enabling CWA to model performance variability. Subsequently, CWA and SAD were applied to road transport to identify performance variability of four different road user groups in intersections. This application confirmed that SAD coped with the added complexity of such a system, and satisfied the criteria for a performance variability modelling method better than the application of CWA without SAD. The application provided a more comprehensive understanding of performance variability in road transport, thereby adding aspects that have to be represented by a performance variability theory and the field of safe travel theory. Next, the ability of CWA and SAD to provide performance variability based evaluations of two different intersection designs was examined. This application demonstrated how CWA and SAD are able to provide a low cost and useful desk top evaluation of performance variability for different road user groups in different road transport designs. It demonstrated how different intersection designs create different levels of performance variability for different road users and how small system changes can have large effects on performance variability. To formally test the new SAD method, the final study involved an evaluation of the reliability and validity of the SAD method when applied by analysts other than the developer. Based on the satisfactory but not optimal performance of the method when used by novice analysts, a group process was proposed and further guidance was developed to enhance the reliability and validity of the SAD. The findings were synthesised into a conceptual framework of performance variability and the field of safe travel theory. The understanding of performance variability possible was advanced, including how variability in performance is shaped by a wide range of factors and high level of interaction within a system. Implications of the research, limitations as well as future research directions were discussed. 

Awards: Winner of the Mollie Holman Doctoral Medal for Excellence, Monash Injury Research Institute, 2013.