"Dribia's knowledge, both algorithmic and implementation of innovative solutions, have been key to the adoption of the project. We are very happy with them."
FEATURES
Daily and monthly traffic volume prediction system segmented by location and typology configurable in different countries.
IMPACT
Prediction errors of less than 2%, prediction error margins and explainability.
ACHIEVEMENTS
Unified but adaptable database and query and modification interface, with API to be able to offer predictive services in a unified way to the units that need it.
BUSINESS PROBLEM
Infrastructure management depends primarily on the expected volume of use. Multiple important aspects to ensure good service (attention, collection, state of the infrastructure, maintenance, etc.) depend on good resource planning. Planning these resources involves establishing in advance the expected volume of users with the maximum possible granularity. This is difficult with traditional tools based solely on business knowledge, which on the one hand make granular data processing difficult, and on the other prevent continuous evolution of models and their improvement.
DRIBIA’S SOLUTION
Dribia has built INTI, a prediction suite that allows unified forecast management, using the same model but adaptable and configurable to each business unit. The tool has a back-end with an API available to all services that need it, thus helping to unify the methodology, data and use of an important source of planning information. On the other hand, there is an interface that allows historical consultation, modification and export of data to facilitate this process, which was previously manual, and revert to greater transparency, ease of use, speed and precision.
