Limitations in achievable performance and programmability are obstacles to realizing productivity gains from the full automation of manufacturing operations requiring a large variety of low-volume tasks. The use of collaborative robots to assist human workers is a promising approach to overcoming these obstacles, without displacing human jobs. Current efforts in this area focus on the worker-robot partnership and overlook the critical role of the supervisor in managing workloads and allocating tasks. By considering the larger context of supervised work teams, this Future of Work at the Human-Technology Frontier (FW-HTF) research aims to enhance productivity and improve worker quality of life by increasing the effectiveness of workers operating in partnership with robots. It provides a framework for analyzing readiness, assessing adoption, and evaluating performance of collaborative robotics in industrial settings. Partnerships and interactions with companies in the Southeastern USA will promote realistic research efforts that translate to practice, benefitting small-to-medium manufacturing companies in the USA. Efforts and findings will be promoted to the public to attract the next generation of workers and researchers to science and engineering fields.
This project explores two hypotheses. The first working hypothesis is that, when workers view robots as partners, imperfection will be tolerated if the worker can successfully manage the robot to complete the task faster than their self-conceived rate. The determining factor regarding the value of a worker-robot collaborative partnership is hypothesized to be the worker?s ability to allocate the task workload between the robot and themselves towards an optimal partnership. The second working hypothesis is that the introduction of a supervisor to guide and promote task allocation will further contribute to enhanced worker-robot performance. This hypothesis builds on the observation that line supervisors interact with multiple workers, and thus are a repository of holistic institutional knowledge regarding good practice. To confirm these hypotheses and arrive at the anticipated framework, called the Worker-Robot Supervisor Effectiveness Model, requires a mixed-methods research design. First, a grounded theory study will establish assessment criteria related to worker, robot, and supervisor technology adoption and performance to develop an instrument for measuring both. Second, within a simulated manufacturing work cell environment, a set of experiments will investigate factors linked to successful technology adoption and work cell performance. Third, the findings from these studies will inform the creation of the model, which will provide guidance for effective integration, adoption, and supervision of worker-robot partnerships. Fourth, to confirm the applicability of the derived model, it will be used to field and integrate a robot within the existing processes of a real-world company. Field deployment will provide empirical evidence to validate and refine the model.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
