Our client is a company of Argentine capital, dedicated to provide geographic information services and AVL systems (Automatic Vehicle Location). It operates more than 20,000 (devices) that include fleets from the productive sector and public service organizations.
One of its clients, the “Police of the Province of Buenos Aires”, had a monitoring system which updated the state of coverage and patrol of the vehicles every 24 hours, thus not being able to make early decisions regarding deviations or incidents. The architecture of the solution was a classic approach to data warehousing based on SQL technologies.
It was necessary to provide a new system, capable of monitoring the patrol coverage in real time, in order to optimize the decision making regarding the geographical deployment of patrollers.
The Solution A platform was developed for the processing of data flows or messages (stream processing), coming from the embedded devices on board the patrol mobiles. In real time, the positioning of mobiles is determined in the maps residing in a graph-oriented database, using high-performance recursive algorithms, which allows computing patrol coverage rates by zones and calculating a series of indicators that carry statistics necessary for a later analysis.
The paradigm of this architecture, is based on the concepts of Big Data and NO-SQL, in its different forms. A farm of servers was used to meet processing requirements of the solution. Likewise, this architecture has analytical components, running machine learning algorithms in order to detect anomalous conditions in patrols.
Finally, the information is shown in two control panels, composed of maps and indicators, which allow to visualize the state of the coverage by location and of the entire province.
The system was put into production successfully without major setbacks, becoming a valuable tool that allows real-time visibility regarding the efficiency and effectiveness of mobile deployment strategies carried out, and take corrective actions to optimize results of prevention and repression of crime.