Individuals working nonstandard hours face a range of negative health, safety, and productivity outcomes, largely driven by sleep disruptions associated with their schedules. Existing interventions to improve sleep often adopt a one-size-fits-all approach, overlooking the diversity of individual needs, preferences, and work contexts. These factors are critical considerations for any intervention aiming to improve the sleep of individuals working nonstandard hours as schedules can differ dramatically, both between individuals and within an individual's schedule. Advances in wearable consumer sleep technology and artificial intelligence, like the use of reinforcement learning and large language models, now offer the opportunity for highly tailored just-in-time-adaptive-interventions (JITAIs), or digital interventions that adapt to individuals' unique contexts to provide personalized, timely behavioral support. This paper proposes that integrating artificial intelligence and wearable consumer sleep technology with JITAIs has the potential to deliver the right support, at the right time, and in the right context for each individual nonstandard-hour worker. By directly responding to the unpredictable and variable hours these workers face, such technologies could set a new standard for personalized health, safety, and productivity interventions. Challenges associated with incorporating artificial intelligence and wearable consumer sleep tracking devices into JITAIs, such as trust, technological and algorithmic inaccuracies, user engagement, and cost, are also discussed as key considerations for successful implementation. This paper is part of the Consumer Sleep Technology Collection.
Smith et al. (Thu,) studied this question.