CHALLENGE

The classic “control cycle” based on monthly closures is limited in timeliness (since it is typically a time-consuming manual process), detail (since it works on aggregate items) and performance (because of the limits of the tools typically used – e.g. MS Excel).

In this scenario, the need arises for a faster, prompt, and timely actual data closing, and for a reliable and frequent predictions on short term closure (during the month).

OUR SOLUTION

BIP supported TIM in developing an automated and scalable platform based on a proprietary Auto-Machine Learning Engine.

Smart Closing allows to manage daily historical data of the main managerial indicators which contribute to the Service Margin, automating the retrieval of all the relevant analysis dimensions (e.g., prices/tariff plans, offers/promotions, customer segments, usage type, …) via RPA or direct DB connections, checking the overall data quality and applying corrections where necessary.

The forecasting engine is based on an Auto-ML service to speed-up the development process and the outputs are presented through Power BI web dashboards that meet all the executive and operative needs.

RESULTS

The solution provided many qualitative and quantitative benefits such as:

• Progressively reducing error in the prediction approaching the end of the month
• Lowering processing times: few hours of computation compared to the 4-5 days previously spent for manual calculations
• User-friendly tool to support visualization of results
• Automated extraction of actual data from Legacy systems through RPA