
- Whitepaper
When Service Becomes the Business Model
How the heavy industry is unlocking new growth potential with service excellence and AI.
From Incident to Value Creation: How Service Excellence Builds Resilience and Recurring Revenue
Between 60 and 80 percent of industrial value creation happens after delivery, in maintenance, spare parts, upgrades, and service-based models. Yet in many heavy-industry companies, service is still seen as a hard-to-plan add-on to the product business. Action is taken only once something breaks down.
Leading industrial companies show that there is another way. Predictive monitoring detects critical conditions before they lead to failure. Field assignments are prepared based on data, and technicians arrive equipped not only with tools but also with the knowledge they need. This turns the individual service case into an end-to-end value creation process. Customers benefit from higher availability, companies from reliable revenue streams, and employees from a noticeably lighter workload.
The whitepaper presents the three levers for a new kind of service excellence. Using the “Referenzlok” (reference locomotive) project at DB Fahrzeuginstandhaltung as an example, we show how to get started pragmatically: with focused use cases, clearly defined processes, and a blueprint that consistently follows the logic of process → data → AI.

Key Questions Addressed in the Whitepaper:
- Why is industrial value creation shifting to the second half of the product lifecycle, and what does this mean for the service business?
- Why is transparency across the installed base the starting point for every service excellence strategy?
- How do the three levers Pre-empt & Monetize, Orchestrate & Deliver, and Empower & Learn improve predictability, first-time fix rates, and service quality?
- How did DB Fahrzeuginstandhaltung use the “Referenzlok” to pilot AI-supported document review and bill-of-materials derivation, and what potential did the pilot reveal?
- Why does service excellence only succeed in the sequence process → data → AI?