How Trackunit built IrisX on Databricks to help reduce equipment downtime
by Domokos Spéder, Erin Kirsten and Jack Yallop Construction generates abundant data, from equipment telemetry and maintenance records to job-site documents and rental feeds. But like manufacturing, the industry faces fragile supply chains, margin pressure, and rising demands for speed, customization, and traceability.
The data remains fragmented across systems, organizations, and equipment types. Ownership can change from project to project, and critical information may be unstructured, disconnected, or never captured. This is more than a data quality problem; it is a data intelligence problem. AI cannot improve decisions when data is never connected, structured, or provided with the right operational context.
Trackunit is addressing this challenge with IrisX, an operating data platform built on Databricks Data and AI Platform. Trackunit’s value lies in its construction-specific context: connecting equipment, machine, operator, site, and operational data across the ecosystem so teams can turn fragmented signals into insights and action. That perspective is grounded in 20 years of industry experience, 5,000 customers, 6 million connected assets across 120 countries, 1,200 connectors, and 150 partner marketplace applications.
By leveraging Databricks, Trackunit can bring data engineering, analytics, and AI together on a single, open platform. This enables IrisX to turn raw machine data into operational decisions and enterprise intelligence through three capabilities: connect, distill, and amplify. Watch Domokos Spéder, VP, Commercial EMEA & Global Consulting at Trackunit, and Erin Kirsten, Manufacturing Senior Solutions Architect at Databricks, share how Trackunit was built on Databricks.
Construction equipment does not operate in isolation. Original equipment manufacturers (OEMs) need visibility into how machines perform after they leave the factory. Rental companies need to understand utilization, availability, and maintenance across locations. Contractors need to know where equipment is working, where it is underused, and where capacity is constrained.
Each organization has different systems and priorities. They may also use different data models and terminology. A machine’s telemetry may live in one platform, service records in another, and job-site information in a document or spreadsheet. Without a shared foundation, even a straightforward operational question can require manual reconciliation.
IrisX brings together data from machines, equipment, operators, documentation, and third-party sources. The platform is designed to work with the tools that construction businesses already use, including tracking applications, enterprise resource planning systems, customer applications, and analytics or AI engines.
This foundation preserves the context behind each signal. A fault code is more useful alongside equipment history, operating conditions, maintenance activity, and location. Utilization becomes more useful when compared with contract terms, project needs, and asset availability.
Connecting data is only the first step. Raw equipment data is not automatically useful to a product manager, service team, fleet operator, or business leader. It must be cleaned, structured, governed, and translated into questions the business can act on.
IrisX applies construction-specific context to that process. In the demo above, a user asks how engine load and torque affect fuel consumption and which equipment cohorts are outliers. The system returns an executive summary, visual analysis, and equipment-type breakdown without requiring a query. A second example examines regional operating hours and seasonal patterns, helping product and engineering teams design for field reality rather than an assumed average.
This change determines who can use advanced analytics, allowing product managers, service teams, and operations users to ask natural-language questions about their fleet and operations and receive answers grounded in governed data. This extended access enables timely intelligence wherever decisions are made.
The same intelligence can also be made available through the tools people already use. Through the Trackunit IrisX MCP, Trackunit can embed IrisX analytics in Trackunit Manager, so users can get answers and take action in their preferred AI tools without switching systems or moving data around.
An insight only creates value when it changes what happens next. That is the purpose of the amplify layer: turn signals into decisions and decisions into action.
This is why Trackunit built IrisX Blueprints, ready-to-deploy solutions for specific construction and equipment use cases. They combine data connections, workflows, analytics, and AI-driven automation logic so teams can start from a business problem rather than a blank development environment. These blueprints are deployable in days rather than months, without custom development, helping teams move from connected data to action faster.
For OEMs, field data can inform product design, service operations, warranty management, and digital offerings. A Battery Management Insights Blueprint can consolidate charging behavior, standby time, battery status, and reporting activity across electrified equipment. Teams can see reporting status and charge levels across the fleet, identify low-charge or inactive assets, and inspect charging sessions over time. The demo above describes a customer who found a pattern of short, shallow charging sessions associated with premature battery degradation. That insight supported better charging guidance, while the same data can inform fleet readiness and battery transparency for resale or buyback.
For an OEM fleet of roughly 10,000 machines produced per year, the battery management blueprint unlocks approximately $3 million in annual value through reduced warranty risk, faster issue diagnosis, and opportunities for digital features. Rental companies need to compare equipment use with the agreed contract terms. An Out of Contract Usage Blueprint can continuously surface overruns and connect them to existing business systems and invoicing processes, replacing weeks of manual reconciliation.
The demo above shows an example in which this approach surfaced roughly $2 million in previously missed invoices for a mixed rental fleet of approximately 5,000 units. It also cites more than 10% higher retention after billing gaps were addressed transparently. Better data can support both revenue protection and stronger customer relationships.
For contractors, fleet imbalance is also a distribution problem. One site may have idle equipment while another faces capacity pressure and turns to emergency rental or an unplanned purchase. A Site Performance & Asset Utilization Blueprint gives every operational leader one view. A connected view of assets, project needs, location, and utilization can identify where equipment should move next. The demo above highlights an example in which data-driven redeployment reduced project delays and overtime and lowered equipment-related costs. When the right asset is already in the business, moving it can be more effective than adding another asset.
The connect, distill, and amplify pattern offers a practical approach to AI adoption.
First, connect machines, operators, sites, service events, contracts, and documents. Second, distill them into governed, contextualized intelligence. Third, amplify that intelligence through applications, workflows, and AI assistants where decisions are made.
This avoids introducing an AI assistant before the data foundation and domain context are ready. In construction, a useful answer must reflect the right asset, site, operating conditions, and business process. The goal is better decisions about service, redeployment, charging, usage exceptions, and product development—not simply more dashboards.
For manufacturers, the same foundation supports operational excellence at scale, resilient and intelligent supply chains, and faster product development. AI can help reduce unplanned downtime, improve yield, cut waste, anticipate disruptions, optimize inventory, and accelerate sourcing decisions before problems reach the bottom line. The common thread is moving from reactive to predictive, from siloed to unified, and from data-rich to genuinely intelligent.
For years, connected equipment has helped construction businesses see more of what is happening in the field. The next step is to use that connected data to improve subsequent decisions.
Organizations do not need to own the largest fleet to operate more effectively. They need a clearer view of their assets, a stronger context around their data, and workflows that help people act on insights while they are still useful.
IrisX, built on Databricks, demonstrates what that operating model can look like: connect fragmented construction data, distill it into intelligence, and amplify it through the applications and processes teams already rely on.
For construction leaders and practitioners, the starting point is practical. Identify a decision where downtime, underutilization, slow service, missed billing, or avoidable asset replacement creates measurable cost. Then work backward to the data required to improve that decision, and build the foundation that allows AI to use it. Ready to see IrisX in action? Watch the full demo and expert deep-dive from our Manufacturing Virtual Industry Forum: From Pilots to Production.
IrisX is an operating data platform built on Databricks that connects construction equipment, machine, operator, site, and operational data, turning fragmented signals into contextual intelligence and workflows.
By combining telemetry with maintenance, project, location, and contract data, AI can help identify downtime, charging issues, missed billing, underused assets, and redeployment opportunities.
Databricks provides one governed data and AI platform for predictive maintenance, equipment utilization, supply chain resilience, quality monitoring, and warranty management—helping teams move from reactive reporting to predictive operations.
Unity Catalog provides a governed foundation for construction data and AI assets, helping teams manage access, lineage, and consistent use of equipment, telemetry, maintenance, and operational data across workflows.
It brings data engineering, analytics, and AI together on one open platform, helping organizations connect machine, site, operator, and operational data and deliver insights through applications, dashboards, and AI interfaces.
Subscribe to our blog and get the latest posts delivered to your inbox.