Sanne Klit: The Data Future: Deliver the Spare Part Before the Failure (Part 4)
(Part 4 of our series with Sanne Egholm Klit, Head of Parts Solutions & Warranty at BEUMER)
What if spare parts could arrive before the failure actually occurs?
For most of industrial history, spare parts management has been a game of "catch-up." Something breaks, the line stops, and the clock starts ticking. It’s a reactive, high-stress cycle that relies on how fast a human can diagnose a fault and how quickly a courier can drive a van.
The reason is simple. Predicting failures requires large amounts of operational data, and until recently that data barely existed. Maintenance records were often fragmented, equipment generated little digital information and analysing behaviour across an installed base of machines was practically impossible.
Today, that situation is changing. Modern equipment produces far more operational data, maintenance systems record interventions more consistently, and advanced analytics can detect patterns across many installations. As a result, predictive spare parts planning, anticipating demand before failures occur, is becoming technically achievable.
In our previous articles with Sanne Egholm Klit, we’ve explored the OEM’s strategic promise (Part 1), the spare parts planning puzzle for OEMs (Part 2), and the brutal reality of obsolescence (Part 3). Now, we’re looking at the final frontier: the transition from "firefighting" to "predictive delivery."
From reactive logistics to preventive support
For most machine builders, spare parts supply has historically been driven by events in the field. When a component fails or approaches the end of its expected lifecycle, operators contact the supplier and a replacement is arranged.
Planning helps mitigate the impact of this uncertainty, but it rarely removes it entirely.
Sanne believes the next evolution of spare parts management will gradually move away from this purely reactive model.
“We would like to deliver spare parts on site before the customer knows that they need them.”
The idea of "predictive parts" isn't actually new. In fact, BEUMER was dreaming about this when the iPad was still a novelty.
“Back in our 2013 strategy we talked about calling the customer before they knew they had a problem.”
So, why has it taken so long? Because, until recently, we were missing the "fuel" for the engine: high-fidelity operational data. A decade ago, maintenance records were often scribbles in a logbook or siloed cells in a spreadsheet. Equipment was "dumb," and comparing performance across different airports or logistics hubs was an exercise in futility.
Today, the "Internet of Things" (IoT) isn't just a buzzword, it’s the backbone of the system. We finally have the sensors.
Moving from "Gut Feeling" to "Field Truth"
For years, OEMs and operators have relied on "Mean Time Between Failure" (MTBF) numbers, theoretical stats from a handbook. But as any reliability engineer like Sanjib Das will tell you, a pump in a humid airport basement behaves very differently from a pump in a dry, climate-controlled warehouse.
Sanne believes the future lies in harvesting real-time data from CMMS systems and sensors to build a "Field Truth."
“We really hope that the data available from sensors and CMMS systems will allow us to become more preventive.”
This perspective is particularly valuable for organisations responsible for supporting large numbers of systems across different customers and locations.
By comparing maintenance data across installations, engineers can begin to identify failure patterns, typical component lifecycles and the conditions that accelerate wear.
The more insight machine builders have into how equipment behaves in the field, the better they can align spare parts availability with real demand.
Why prediction changes the relationship
Predictive insights do more than improve forecasting accuracy. They change the relationship between machine builders and operators.
Today, spare parts supply typically begins with a request from the operator. A failure occurs, a component is identified and the supplier is contacted.
In a more predictive model, the flow of information can gradually shift. Machine builders may be able to detect patterns across multiple installations before operators experience the same issue.
When that happens, spare parts supply becomes less reactive and more proactive. Replacement components can be produced, stocked or positioned closer to the customer before failures begin to affect operations.
For operators, this reduces the risk of unexpected downtime. For machine builders, it strengthens their role as long-term partners in system reliability.
The challenge of shared visibility
Despite this potential, predictive spare parts management is not something any single organisation can achieve on its own.
The data required to make it work is distributed across the ecosystem.
Operators hold detailed operational and maintenance records. Machine builders hold engineering knowledge, component structures and visibility across the installed base.
Bringing these perspectives together remains one of the central challenges of the next phase of industrial service models.
As Sanne puts it:
“Today we still rely a lot on knowledge and gut feeling. But we are missing the data.”
Bridging that gap will require both technical integration and closer collaboration between operators and machine builders.
Towards a more connected spare parts ecosystem
The gradual shift toward data-driven spare parts management is already influencing how organisations think about inventory and availability.
Instead of optimising spare parts purely within the boundaries of a single warehouse or site, companies are increasingly exploring more connected approaches to inventory management.
Shared visibility across locations, coordinated stocking strategies and collaborative planning can significantly improve availability without dramatically increasing inventory levels.
Solutions such as SPARROW.Stock and SPARROW.Pool help manufacturing organisations explore these approaches by optimising spare parts availability across multiple sites and enabling shared inventory strategies where appropriate.
In such models, spare parts are no longer managed as isolated inventories but as part of a broader reliability network.
What this series reveals about spare parts
Looking across the conversations in this series, a few patterns become clear.
Spare parts management is no longer a purely operational activity. It is a strategic capability that sits at the intersection of engineering knowledge, supply chain coordination and long-term customer relationships.
Machine builders must plan inventory across decades of equipment lifecycles. Operators must balance uptime requirements with the capital tied up in stock. Suppliers must manage their own evolving product portfolios and component lifecycles.
In that environment, spare parts planning can no longer rely solely on experience or historical demand. It increasingly depends on data, transparency and collaboration across the entire ecosystem.
Seeing spare parts differently
Across this series with Sanne Egholm Klit, one theme has appeared repeatedly: spare parts management is fundamentally about relationships.
Machine builders support systems that may remain in operation for decades. Operators rely on those systems every day to keep their operations running. Suppliers must ensure that components remain available even as technologies and product generations evolve.
Spare parts sits at the intersection of these responsibilities.
When the system works well, it remains largely invisible. When it fails, the consequences are immediate and costly.
But as Sanne’s perspective shows, the future of spare parts management may look very different from its past.
Instead of reacting to the next failure, organisations may increasingly be able to anticipate it.
And in that future, the most successful machine builders will not simply supply spare parts.
They will help ensure that the right parts are already available when the system needs them.
Read the full series
1️⃣ Spare Parts Is Not a Side Business for Machine Builders
2️⃣ You Cannot Stock Everything. But You Are Expected to Know What Breaks Next
3️⃣ The Obsolescence Trap: Why “Proactive” Is Harder Than It Sounds
4️⃣ The Data Future: Deliver the Spare Part Before the Failure
