Coordinating the interruption of assembly workers in manufacturing

Robot Classification of Human Interruptibility and a Study of Its Effects

“…This discrepancy is explained in part by the results from HFE research (Section 2.2), which suggests the embodiment of the robot interruptions and the skill-based main task contributed to unaffected task performance: it is likely that participants were able to optimize their build process such that their performance remained unaffected on the metrics of task throughput that we instrumented. With better instrumentation, future research has the potential to examine additional metrics of task performance, such as interruption resumption lag [30,33,43], which should differ between RND and MDL according to the predictions of the Goal-Activation model [2].…”

“…Similarly, in embodied settings, when people defer an incoming interruption, they are more likely to complete their original task [23]; a result predicted by Prospective Memory [38] models of interruption handling [24]. Recent results from HFE continue to show that performance loss is not noticeable with tasks that are embodied or skill-based, even when the interruptions might be computer mediated as in the work of Lee & Duffy [35] and Kolbeinsson et al [33]. These authors, in particular, reason that performance loss is absent in an embodied setting because it is impossible to occlude the main task, which allows people to optimize common sub tasks and choose when to switch to an interruption.…”

“…In HCI and HFE, the cost of an interruption on-screen has been evaluated with quantitative metrics such as time on task [1,33,37,40], the number of tasks completed [40], the number of incomplete tasks [23], the number of errors [35,40], switching time [30,33,43], and workload [1,37]; and qualitative metrics such as respect [1] and preference [40]. In embodied settings, researchers have also used structured interviews [24,55,57] and ethnographies [18,24,45,55] to evaluate long-term interruption costs.…”

“…Prior work in HCI and HFE quantifies task performance using metrics such as time on task [1,33,37,40], the number of tasks completed [40], and task switching time [30,33,43]. We use similar quantitative measures of human and robot performance, and 5-point Likert scale responses to questions of participant opinions and participant background: interruption lag, measured as the time between when the robot requests assistance and the participant begins constructing the robot build; interruption duration, measured as the total time the robot waits after it has requested assistance.…”

Interruption Cost Evaluation by Cognitive Workload and Task Performance in Interruption Coordination Modes for Human–Computer Interaction Tasks

“…Arbitrary, contrived tasks can manipulate design factors such as task difficulty and interruption moments and durations, whereas actual tasks in natural settings are easily supported by various post-interruption tools and artifacts, which guide one back into the task in other ways. Several interruption studies were carried out with actual authentic tasks including the comparison of interruption modes in a manufacturing process [66] and the investigation of resumption lags in interrupted critical care tasks in the healthcare domain [67]; however, more interruption studies are still needed to understand how to minimize the negative effects of interruptions.…”