Paper:
Noa Segall, Rebecca S. Green, and David B. Kaber. 2006. User, robot and automation evaluations in high-throughput biological screening processes. In Proceedings of the 1st ACM SIGCHI/SIGART conference on Human-robot interaction (HRI '06). ACM, New York, NY, USA, 274-281.
Review:
This paper explains a task analysis that was done to high-throughput screening (HTS) system in the field of biological samples of life sciences. The analysis in this paper was done to an existed HTS system at the Center for Life Sciences Automation (CELISCA) in University of Rostock (Germany). This type of systems include potential human errors as human operator has to plan the procedure before the process. Those errors are costly, that's why authors have conducted this work. This paper has two goals: (a) minimize the human involvement to monitor the system errors involved (automation) , and indentifying the limitations of the current interface design to help in automating the heavy manual workload.
This work focuses on one part of the HTS process which is human-robot interaction (HRI) as its errors are costly. In HRI, there are 11 steps needed to be done by the operator (see table 1). Most of those steps are human. As a result of that, human errors can occur. Additionally, some chemical errors might happen.
In order to address the problem of the costly errors in the human-robot interaction (HRI) in the HTS process, authors used cognitive task analysis (CTA) to facilitate the operations of HTS. In this context, they applied goal-directed task analysis (GDTA) jointly with abstraction hierarchy (AH) modeling as a new integrated CTA approach (see section 4). GDTA were used to identify the current tasks and their workflow, while AH were used to automate those tasks on the user interface.
For the GDTA, authors used it to identify users’: major goals, subgoals, operational tasks, questions about decision making in task performance, and developing information requirements. They elicited the tasks using structured interviews with one domain expert at the university. Two examples of GDTAs can be found in figure 3, and 4. Later, they used the results of GDTA to generate the abstract hierarchy (AH). They gave one example of abstract hierarchy (AH) for the barcode device which is used in figure 3 for printing.
The recommendations of this study were to enhance the user interface by bringing more automation to the current interface and therefore eliminate some of the potential human errors. One example of those recommendations is having a default labels for barcodes with the option of user defined barcodes.
What I liked about this paper is the combination of both techniques GDTA along with AH to serve the purpose of this study, however I would suggest to them a further investigation in validating their recommendations by experimenting them with HTS users. Another issue with this paper is that it has many acronyms which makes confusing to readers.
Noa Segall, Rebecca S. Green, and David B. Kaber. 2006. User, robot and automation evaluations in high-throughput biological screening processes. In Proceedings of the 1st ACM SIGCHI/SIGART conference on Human-robot interaction (HRI '06). ACM, New York, NY, USA, 274-281.
Review:
This paper explains a task analysis that was done to high-throughput screening (HTS) system in the field of biological samples of life sciences. The analysis in this paper was done to an existed HTS system at the Center for Life Sciences Automation (CELISCA) in University of Rostock (Germany). This type of systems include potential human errors as human operator has to plan the procedure before the process. Those errors are costly, that's why authors have conducted this work. This paper has two goals: (a) minimize the human involvement to monitor the system errors involved (automation) , and indentifying the limitations of the current interface design to help in automating the heavy manual workload.
This work focuses on one part of the HTS process which is human-robot interaction (HRI) as its errors are costly. In HRI, there are 11 steps needed to be done by the operator (see table 1). Most of those steps are human. As a result of that, human errors can occur. Additionally, some chemical errors might happen.
In order to address the problem of the costly errors in the human-robot interaction (HRI) in the HTS process, authors used cognitive task analysis (CTA) to facilitate the operations of HTS. In this context, they applied goal-directed task analysis (GDTA) jointly with abstraction hierarchy (AH) modeling as a new integrated CTA approach (see section 4). GDTA were used to identify the current tasks and their workflow, while AH were used to automate those tasks on the user interface.
For the GDTA, authors used it to identify users’: major goals, subgoals, operational tasks, questions about decision making in task performance, and developing information requirements. They elicited the tasks using structured interviews with one domain expert at the university. Two examples of GDTAs can be found in figure 3, and 4. Later, they used the results of GDTA to generate the abstract hierarchy (AH). They gave one example of abstract hierarchy (AH) for the barcode device which is used in figure 3 for printing.
The recommendations of this study were to enhance the user interface by bringing more automation to the current interface and therefore eliminate some of the potential human errors. One example of those recommendations is having a default labels for barcodes with the option of user defined barcodes.
What I liked about this paper is the combination of both techniques GDTA along with AH to serve the purpose of this study, however I would suggest to them a further investigation in validating their recommendations by experimenting them with HTS users. Another issue with this paper is that it has many acronyms which makes confusing to readers.
No comments:
Post a Comment