Natural Resources Company
Haze Software was engaged to provision an analytical platform that would enable intuitive near-real time reporting for the asset management function of a natural resources organisation.
The challenge
Key challenges:
- near real-time reporting on operational and financial datasets from transactional systems
- an intuitive interface for exploratory and structured self-service reporting
- supporting both long-term trend analysis and ‘live’ operational reporting
- scalable to handle significant volumes of operational data
- ensuring teams can only access data relating to their own operations
- minimising platform operational costs through automation.
The approach
The engagement followed three high-level phases: discovery, design and build. Each phase consisted of a set of specific activities and resulted in a set of tangible outcomes.
It was critical to obtain a clear understanding of the platform requirements and the current state architecture of the data landscape. A workshop was conducted with key stakeholders during the discovery phase to facilitate this process.
During the design phase, the team executed a series of ‘proof-of-concepts’ with a goal of mitigating specific technical risks and informing any significant architectural decisions. These activities and a technical workshop with stakeholders delivered the primary output of the design phase, the target-state architecture.
An agile framework was used during the build phase to follow the principle of incremental delivery of value. Foundational elements of the platform were established early and then an asset performance data model was migrated to the platform as a high-priority use case while the remaining functionality was built-out.
Documentation for the as-built platform architecture and the platform operational procedures were produced for the client to enable in-house maintenance by subject matter experts.
The outcome
Key outcomes included:
- ‘right-sizing’ of platform infrastructure and configuration for optimal performance
- near-real time exploratory reporting on operational and financial data sets
- standardised historical and ‘live’ reporting of asset performance
- a fully-automated process for incrementally loading data in near-real time using a rolling-window pattern
- a low-maintenance row-level security implementation that ensured that teams could only access data related to their own operations
- integrated logging with an existing enterprise monitoring application.