UTASiMo: a simulation-based tool for task analysis

Abstract

Research on task analysis and human performance has focused on the development of adequate tools, models, and methods to understand, analyze, and improve the relationship between humans and systems. As technology continues to advance and to change the nature of human work, techniques of analysis are changing to meet the new needs. This work attempts to fill the gaps in the current task analysis tools and describes the architecture and development of a simulation model, named UTASiMo. UTASiMo is a simulation tool that aims to enhance task analysis by automatically generating a multi-method simulation model for well-defined tasks based on a spreadsheet template filled in by the user. The generated model analyzes and simulates tasks performed by individual simulated agents representing human operators while accounting for the estimation of an operator's utilization and error prediction. This work highlights the design, development, and evaluation of the hybrid architecture (discrete event and agent based) of UTASiMo. The development of the system dynamics model, which is responsible for the human error assessment, is a work in progress and is excluded from the present paper. A real-world case study has been adapted to evaluate the hybrid architecture of UTASiMo. The same case study was modeled using the Micro Saint simulation tool. The results produced by UTASiMo were compared with the real-world data as well as with the results produced by Micro Saint for validation purposes. The comparisons indicate the validity of the UTASiMo-generated model and also that the hybrid architecture produces more variability in the results than using only one method. The comparisons also show promise that the tool will reduce the time and effort of the task analysis simulation.

A modeling framework for the application of multi-paradigm simulation methods

“…62 For sub-objective o 2 , the framework recommended DE to capture the temporal aspects of tasks, and for sub-objective o 3 , the framework recommended AB to capture the human heterogeneity, behavior, and actions. 63…”

“…The hybrid architecture was suggested by the interaction point (Human Error (SD) – Agent (AB)) in section 4.3. Human operators were modeled as agents using a state-chart with three states (Figure 17) 63 :…”

“…More information about the V&V process and documentation of results can be found elsewhere. 60, 62–64…”

A hybrid simulation model for crowdsourced software development

“…1) Macro-level Competition Model: At the top platform level, the SDS model, is applied to represent the overall relationship amng task arrival, worker reliability, and task competition outcome by revealing a pattern of broader market behavior [27] [28]. As illustrated in macro level of Figure 2, it contains 14 variables including task, agent decision, workers' performance, task similarity, worker profile, worker skillset, and different available crowdsourced markets.…”

A simulation model for estimating human error probability

“…The current paper focuses on the description of the SD module of a "work-in-progress" hybrid simulation model for the estimation of HEP by utilizing the SPAR-H method's Performance Shaping Factors (PSFs). The model is a continuation of our previous work (Angelopoulou 2015;Mykoniatis 2015;Angelopoulou and Mykoniatis 2017;Angelopoulou and Mykoniatis 2018;Mykoniatis and Angelopoulou 2019). The remainder of the paper is organized as follows: Section 2 provides an overview of the SPAR-H human reliability assessment method and the defined PSFs, while Section 3 describes the SD architecture of our hybrid AB-SD model and the cause and effect relationships of the SD model components.…”