How we did it
This was an exploratory project and it was important that the approach demonstrated value during each phase. The agile approach ensured the project could pivot quickly when a new opportunity emerged.
It was necessary to develop deep empathy for the business operation, how the current systems supported the real-world process and how users interacted with the core applications. This understanding enabled the project team to better interpret both the underlying data and how to approach the analytic questions. To enable the organisation to look to the future we constructed a ‘time machine’ which enabled us to go back to any point in time to view key KPI’s. Predictive analytics were then deployed to report future performance. We were also able to go to any point in time to ascertain how accurate past reporting had been.
To improve both data quality and process compliance we developed a ‘digital twin’ of the core processes. We also needed to extract data from Network Rail’s core systems, denormalise it and prepare it for advanced analytics (AWS and Databricks were selected) To support business reporting a Redshift/PowerBI solution was implemented.