Figure 1: Worker with intellectual disabilities assembling an electrical cabinet in collaboration with a cobot. The robot checks the cables the worker has just connected and informs the worker if an error is found, giving them an opportunity to correct it. Automation technologies are transforming work, and this affects workers with any profile, including workers with disabilities, for whom technologies create new and exciting employment opportunities. In this position paper, we outline our research work on the introduction of some such technologies in real industrial production environments where assembly workers are persons with cognitive disabilities. We focus on collaborating with a robot that complements the workers in cognitively demanding tasks in an assembly job. We also report the ease with which workers with cognitive disabilities understood and could use virtual reality for training programs. As a main message, we defend that design-for-all principles, as used in our work, result in more usable systems and in better designs from which every worker can benefit, regardless of disabilities. CCS CONCEPTS • Human-centered computing • Accessibility • Accessibility systems and tools Additional Keywords and Phrases: collaborative robot, virtual reality, augmented reality, worker with cognitive disabilities, factory of the future, automation, work satisfaction
Automation is transforming work for every person in society, including for workers with disabilities. Taking a
social inclusion stance, persons with cognitive disabilities have the same right as any other member of society
to access fulfilling jobs [
With the growing transformation of work through automation, an opportunity opens up to make workplaces
accessible as never before, if the right approach to design is taken. What is more, when effort is invested in
designing automated workplaces following the principles of design-for-all [
Disabilities can be motor, sensory, mental or cognitive (often a combination of some of these) [
Currently, industrial organizations providing sheltered employment [
An underlying question was whether WCDs could adapt to some of the new production technologies, understand them, and tolerate them in the workplace, or even benefit from them and be willing to adopt them. For that, a set of technologies were analyzed, tested, and in some cases used in realistic production processes (replicas of real processes).
The technologies included were: Collaborative robots, Augmented reality (projected on the workspace or obtained through head mounted devices), Fully immersive virtual reality, and Exoskeletons.
For each technology and for the extent of the evaluation that was conducted with it, stakeholders were involved. First, an expert evaluation committee was named, which included members with most of the following profiles: Support technician specialized in production processes; Support personnel on training and job induction processes; Management responsible for each production process; Clinical psychologist; Ergonomist; Expert in the adaptation of production processes for persons with disabilities; Expert in prevention of health and 1 http://www.manuwork.eu/ safety hazards; Interaction design experts; Technologists (experienced providers of the automation technologies).
In addition, assembly workers (WCDs) were recruited for each study. A mix of worker profiles with different
levels of experience and capabilities to carry out the intended assembly work were recruited.
The research effort aimed at recording the experience with the technology at three points in time:
• Before encountering the technology. Round table discussions were held between the participants
(WCDs) and the expert committee. During these, prior knowledge and experience, or awareness about
the existence of the technology were discussed and recorded. Participants were asked to express in
drawing (Figure 2, Right) what they expected was going to happen and what the experience would be
like (e.g., if a robot, what it would look like, what size, with or without legs, head, talking or not, male
or female…). Worker satisfaction of WCDs and support personnel was assessed with an ad hoc
questionnaire.
• During the interaction. All reactions and interactions were recorded for later analysis (e.g., video
capture, note taking by the observing expert). Think aloud from the participants was encouraged by a
trusted facilitator that dialogued with them. In some cases, standard post task questionnaires (e.g.,
SEQ [
One of the more challenging assembly jobs for WCDs is the wiring of electrical cabinets. Although these workers are normally very effective at connecting cables with a screwdriver, they find a barrier in having to connect them according to documents with electrical schematics, which they cannot interpret. For this reason, traditionally only WCDs with the highest cognitive capacity could opt for this job.
In partnering with a robot, the traditional distribution of tasks, where the human should undertake problem
solving in all its forms, was not viable. In our intervention [
As a result of this intervention, the percentage of all WCDs in the organisation that were eligible for this job grew from 15% to 85%. This was also recorded as a significant boost in confidence, satisfaction and perception of prestige in their organisation, as expressed by many participants.
We learned that it was essential that the robot preserved the level of knowledge and initiative of workers, never attempting to relegate them to a second level by imposing specific ways of doing things. As workers became proficient, the robot had to step back and limit itself to quality check routines.
One of the primary roles of virtual reality (VR) in the factory of the future is the training of human workers in new jobs and procedures. We conducted training sessions with WCDs who had not previously used VR. We found that the behaviour and reactions of workers with a range of disabilities were comparable to the reactions observed in the general population. The more challenging experiences in fully immersive environments were related to teletransportation, although most learned to understand it after dedicated one-to-one instruction. In some cases, the interaction of real objects and people that did not appear in the virtual scenario was reported to be disconcerting. This means that such external objects need to be introduced carefully to begin with. We also observed great resourcefulness from a participant on a motorized wheelchair (Figure 4, left), who did not have difficulties when leaving one of the controllers on her lap (and thus the hand visualization in the virtual world) to have her hand free in the real world and move her wheelchair around.
From the research we have conducted with WCDs and technologies for the factory of the future, we have consistently observed that the reactions and attitudes these users have to the technology and to its uses in work contexts are similar to those observed in the general workforce. A reason for this good acceptance seems to be that WCDs are young on average, and technologically literate (e.g., they are active in social networks with their smart devices). Thus, technology was perceived as friendly and exciting, not threatening as it is sometimes portrayed in popular culture.
We saw that, also with WCDs, preserving control and a sense of agency is very important. Multimodal interfaces should be redundant in the information they present, and the user should be able to attend to the channels of choice without missing anything essential.
Many of the limitations of WCDs are incidentally found in the regular workforce, and systems designed for all have the potential to benefit every user.
The MANUWORK project has received funding from the European Union’s Horizon 2020 research and innovation programme, under grant agreement No. 723711.