Suspensys case

Lakehouse modernization with generative AI on Databricks for scalable, governed, and faster factory analytics

Suspensys, one of the world’s leading manufacturers of suspension systems, axles, hubs, and brake drums for commercial vehicles, trucks, and trailers, partnered with DB to modernize its analytics architecture and move from a fragmented data environment to a governed Lakehouse on Databricks. 

The company relied heavily on Microsoft Fabric and Power BI dashboards in Import mode, which created performance bottlenecks, latency, and maintenance overhead as data volume and complexity increased. Critical information was distributed across SAP S/4HANA via DataSphere, CSV files, legacy systems, APIs, and non-integrated Data Warehouses, limiting reuse, increasing integration effort, and concentrating access to insights in a small group of technical users.

DB designed and implemented a Databricks Lakehouse architecture to centralize data, improve governance, enable low-latency consumption through DirectQuery, and prepare the platform for generative AI, natural language queries, demand forecasting, and multi-plant expansion.

The solution unified corporate data sources into a scalable and governed foundation, structured across Bronze, Silver, and Gold layers. It introduced centralized access control, auditability, end-to-end traceability, CI/CD practices, secure integration with corporate standards, and Databricks AI/BI capabilities to accelerate insight generation for factory operations.

With Genie and natural language analytics, business teams such as PCP, PCM, Materials, Production, Logistics, and Controllership gained self-service access to operational data. Insights that previously took up to two hours became available in minutes. Forecast analysis moved from manual cycles involving senior analysts to a continuous automated flow with SAP integration and faster market-variable input. Performance indicators shifted from manual extraction and validation routines to instant, automated access with higher data reliability.

The modernization eliminated Power BI Import bottlenecks, reduced pipeline execution time through end-to-end automation, removed spreadsheet-version failures, and cut business-area maintenance effort from 12 hours per month to zero. The Gold layer also enabled cost transparency and customer-level allocation rules, strengthening governance and financial control.

The platform now supports higher data complexity, concurrent queries, ad hoc analysis, and expansion across new plants and business domains. The 2026 roadmap includes Spark Declarative Pipelines, intelligent agents integrated with the Lakehouse, demand forecasting, simulation, automated routines, and rollout across Planning Logistics, PPCPM, Operational Logistics, Production, Maintenance, Quality, Controllership, Commercial, and Engineering.

With support from Databricks architects, DB helped Suspensys establish an enterprise Lakehouse standard for Randoncorp: governed, scalable, AI-ready, and designed to support faster tactical and operational decisions across the production chain.