The OEM perspective: Managing what you delivered
For service-intensive Original Equipment Manufacturers (OEMs) building products like construction machinery, heavy industrial equipment, or advanced B2B technology solutions, the core challenge is maintaining visibility over machinery you built but no longer physically control. To protect service margins, hit strict availability SLAs, and deploy new data-driven services, you need clear insights into how your equipment is used and how it performs in the field. Yet, because each client operates your machinery differently, tracking the operational characteristics and equipment modifications over their lifetime remains an ongoing struggle.
The owner-operator perspective: Managing what you run
For owner-operators running production sites in the process industries, utilities, or heavy infrastructure, the reality is defined by a mixed-fleet environment. Inside a single refinery, chemical plant, or paper mill, you run critical machinery from dozens of different manufacturers, purchased across different eras, running on varying hardware and software configurations. Your mandate is to maximise plant uptime and control total cost of ownership (TCO), but establishing a standardised way to monitor and maintain this diverse layout across the entire enterprise is incredibly difficult.
The common bottleneck
The business entry points differ. The OEM wants to optimise field service execution and unlock customer-facing value, while the operator wants to protect plant availability and minimise operational costs—but both hit the exact same data wall.
True operational context is missing because managing industrial equipment requires bringing three completely different data domains out of their isolated silos:
- Installed base master data and topology: Before you can draw operational insights or allocate activities against your technical inventory, you need to shape your concepts and context. This means structuring the composition of industrial systems in a unified way that mirrors physical realities. To be useful, this metadata must be standardised into a common, source-system-agnostic format rather than remaining siloed within separate legacy applications.
- Operational telemetry & time-series streams: Once the physical equipment compositions—whether as-built, as-commissioned, or as-maintained hierarchies—are defined, the next step is using these hierarchies as the technical context for operational telemetry streams from the equipment and its sub-systems. This is where high-frequency machine telemetry, sensor streams, and IoT analytics come into play. Combining physical context with live telemetry streams provides the baseline foundation for calculating critical operational metrics like availability, utilisation, and equipment uptime.
- Maintenance transactions & asset history: Finally, you must track every physical intervention over the equipment’s lifespan. This means linking transactional records directly back to the specific asset, such as work orders, inspection logs, service tickets, component rotations, and physical dispositions. Tracking this transaction history is the only way to measure actual labour and spare parts consumption, making it essential for calculating true lifecycle costs.
The goal: A true digital twin foundation
When you anchor these three elements together, the technical master data, the operational telemetry, and the maintenance history, you have essentially built a digital twin foundation. You move away from static spreadsheets and blind spots, gaining a complete, living record of an asset’s past, present, and predicted future.
The challenge is that connecting these three domains manually is a major architectural hurdle. Telemetry data is high-volume and unstructured, while ERP and CMMS data are rigid and relational. Building custom data pipelines from scratch takes months, and they tend to break whenever a firmware version changes or a maintenance workflow is updated.
The architecture: Databricks as the engine, IBF as the accelerator
Solving this requires a highly capable data platform paired with a ready-made structural framework. That is where the Databricks Data Intelligence Platform and Solita’s Installed Base Foundation (IBF) come in.
Solita’s Installed Base Foundation is the catalyst that cuts down your time-to-value on that platform. IBF is not a closed-box software product or a restrictive SaaS application. It is a Databricks Brickbuilder accelerator that delivers a pre-built reference architecture and a core reference data model designed specifically for heavy asset environments.
We have combined the reference data model with a reference architecture for Databricks to accelerate the bring-up and development of applications combining real-time and historical data into operative systems. Real-time, batch and alert data are read to the Databricks Lakehouse data management system using the Databricks Lakeflow data pipeline solution. In Lakehouse, the data is aggregated and enriched with predictions and forecasts. This data is then moved to Databricks Lakebase, a serverless Postgres database, for low-latency data servicing. Additionally, sending real-time data to Lakebase via the operational system enables seamless combining of historical data, forecasts and real-time data for operative applications.
Instead of spending months whiteboarding complex asset hierarchies, IBF gives you the ready-made blueprint to ingest, clean, and map your master data, live telemetry, and historical service records into an industry-standard information model. It skips the foundational trial-and-error, allowing faster time-to-value.
The business case: Why do this now?
By accelerating your mixed-fleet data consolidation, you unlock immediate, concrete value across your industrial operations, maintenance, and service execution:
Optimised service & logistics planning: Make informed decisions based on physical footprint, asset installations, and their operational characteristics to understand exact service requirements and simplify lifecycle costing.
Lifecycle service innovation and sales excellence: Discover clear opportunities from the existing installed base to grow OEM after-sales and services. Use installed base insights as direct inputs for product design and engineering to develop more competitive solutions.
Performance and availability analytics: Consolidate operational telemetry feedback loops and service lifecycle data to develop value-adding data services for both internal and external parties.
Field service execution: Leverage a single source of data from industrial installations for technical analysis, troubleshooting, and O&M planning to empower technical support and service organisations.
Standardised, governed, and scalable
For IT and data engineering teams, the Installed Base Foundation removes the burden of building custom integration pipelines from scratch for every new equipment variant or acquired asset. You deploy a standards-driven, scalable industrial data architecture directly into your own Databricks environment. The data remains completely yours. Fully governed, secure, and ready for advanced machine learning or generative AI diagnostics when you are.
Managing a distributed fleet requires a permanent, scalable data strategy. You need a powerful enabler platform and a pragmatic blueprint to get you there fast.
Ready to bridge the gap between your machine telemetry and service history? Reach out to our industrial data team, or contact your local Databricks Account Executive to learn how the Installed Base Foundation can accelerate your time-to-insight.
