Send message to Author

Paper short abstract

Predictive Maintenance (PdM) promises to automatize industrial repair, yet its adoption remains uneven. This paper examines the innovation promise of PdM, focusing on data infrastructures, neglected worker expertise, and its embeddedness in future imaginaries.

Paper long abstract

Predictive Maintenance (PdM) has long been heralded as a cornerstone of industrial digitalization, promising to facilitate repair by initiating interventions before breakdowns occur. Yet, despite its conceptual appeal, widespread adoption has faltered: PdM oftentimes seems to be overly complex, unprofitable or unmanageable. The recent proliferation of standardized data infrastructures, industrial platforms, and most recently agentic AI systems has reignited interest in PdM. Are these the missing pieces for its final “breakthrough”?

This paper examines the innovation promise of PdM through three interconnected lenses: infrastructural foundations, datafication of knowledge and work, as well as its embeddedness in future imaginaries.

First, PdM is a method of repair that heavily relies on the integration of heterogeneous data. Current efforts in industrial platformization try to enable standardized aggregation and analysis of data across organizations, but also introduce new dependencies and vulnerabilities.

Second, PdM depends on neglected worker expertise: While maintenance is framed as a data-driven, autonomous process, it often builds on tacit knowledge of workers. How do these infrastructural and organizational aspects shape the possibilities and limits of automated repair?

Third, I conceptualize PdM as a fictional expectation (Beckert 2016). It embodies visions of seamless, efficient, but also sustainable industrial futures. By analyzing how these are constructed and contested, the paper asks: Whose futures does PdM enact, and what ambiguities and alternatives are downplayed or obscured?

As a specific site of sociotechnical transformation that aims on automatizing repair, I argue, PdM underlines how collective imaginaries of digitalization reconfigure organizational practices and redistribute expertise.