Navigating employee–robot collaboration in service contexts: the paradoxical nature, mechanisms, and outcomes

Abstract

While employee–robot collaboration (ERC) is rapidly reshaping frontline service delivery, the field lacks an integrated understanding of how collaboration unfolds, through what mechanisms it produces adverse employee behaviours, and under what configurations it yields desirable outcomes. This thesis addresses these gaps through three interrelated studies that collectively advance knowledge of the paradoxical nature, psychological mechanisms, and configurational pathways of ERC from the employee perspective.

Study 1 conducts a hybrid systematic review of 78 empirical and conceptual articles to map the ERC landscape. Synthesising findings across pre collaboration, during-collaboration, and post-collaboration stages, it develops a processual framework that reconciles contradictory findings by revealing how enabling and constraining forces coexist throughout the collaboration process. Integrating paradox theory with an interdependence perspective, it further proposes a typology of four ERC modes — Guided, Coordinated, Autonomous, and Collaborative — defined by robot autonomy and employee–robot integration.

Study 2 shifts attention to the dark side of ERC. Drawing on mind perception theory and psychological ownership theory, a survey of 332 hotel employees in China examines how collaboration intensity triggers service sabotage directed at customers. The results reveal a serial mediation pathway through perceived agency and self-identity threat, while collective psychological ownership attenuates the link between identity threat and sabotage.

Study 3 adopts a configurational approach using fuzzy-set qualitative comparative analysis with time-lagged data from 253 frontline employees across diverse service sectors. Grounded in an integrated interdependence framework, it identifies multiple equifinal pathways to high service efficiency and employee well-being, yielding a strategic typology of three collaboration archetypes: interdependence-context-dominated, robot-characteristics dominated, and paradoxical-synergy.

Together, the three studies offer a multi-level account of ERC, theoretically by bridging fragmented perspectives into an integrated framework, methodologically by combining systematic review, structural equation modelling, and configurational analysis, and practically by equipping service organisations with diagnostic tools to design, manage, and sustain effective human–robot frontline teams.

Date of Award	19 Jul 2026
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Yangyang Jiang (Supervisor) & Yi Wang (Supervisor)