Using the cloud and artificial intelligence resources, the agency implemented a system of historical and predictive analysis to improve the management of water resources in the region and avoid rationing in periods of drought.
Responsible for the management of water resources in the Federal District, including the reservoirs that supply the population, Adasa (Agência Reguladora de Águas, Energia e Saneamento do DF) has a complex challenge: managing water uses and levels to avoid conflicts in periods drought or even a new rationing. To deal with this scenario, Adasa recently started to rely on the support of Artificial Intelligence in the Microsoft cloud to monitor scenarios and anticipate possible water crises in the DF.
To make a predictive analysis of the next dry periods, Adasa has been using sensors installed at different points to measure water levels in rivers and reservoirs, as well as rainfall rates, with measurements scheduled every 15 minutes. This information is then transmitted via satellite to Adasa's SQL Server Database, installed in the Microsoft Azure cloud.
Subsequently, this data is automatically processed and transferred to various Analytics Microsoft Power BI Dashboards, Microsoft's data analysis tool. Through Power BI, it is possible to perform programmed analyses, such as comparing measurements collected in real time with data from a historical series with more than 30 years of records. This allows the assessment of contexts, perception of trends and the projection of possible behavior scenarios, providing, for example, the anticipation of periods of water restriction. Thus, specialists have access to complete and interactive reports on each reservoir and certain rivers updated on an hourly basis, including the issuance of alerts if the data indicate any risk of conflict over water use.
When everything started
Since 2009, Adasa has installed sensors in the field to monitor rainfall, flow and levels in the Federal District's reservoirs. However, until 2017, you only had access to the data when the team went to the field, every two or three months. The installation of 16 sensors with satellite data transmission, associated with the use of Power BI and cloud resources, allow today not only to visualize the data in real time, but also based on the history of rainfall and flows. region, it is possible to carry out predictive analyzes capable of supporting decision-making even before the dry season in the DF. According to Geraldo Alves Barcellos, responsible for the information technology service at Adasa, “before starting to use Power BI, this history of more than 30 years of data was less used by the Adasa team, as it was not a automatic process, its comparison with new measured data and its use as a basis for scenario prediction.” With the experience gained, another 30 automatic stations with satellite data transmission are being installed by Adasa.
In addition to supporting the work of managing and monitoring water resources, the information generated is made available to the public through the Federal District Water Resources Information System (SIRH), available on the Adasa website. Since the SIRH was implemented in 2017, the number of accesses recorded until July 2019 was approximately 5 million, with these accesses coming from all continents, totaling 99 countries. At one point, it reached 20,000 hits per day.
“Cases such as Adasa demonstrate that the Public Sector has an enormous wealth of data that, through technologies such as the cloud and Artificial Intelligence, can be used to improve resource management on the most diverse fronts. We believe in the expansion of projects like Adasa, reaching other regions of Brazil and other areas where technology can benefit the population”, says Djalma Andrade, vice president of sales for the Public Sector at Microsoft Brazil. Currently, Adasa concentrates all of its servers hosted on Microsoft Azure, while Office 365 is used in the daily routine of the institution's teams.
For the director of Adasa, Jorge Werneck, these and other tools have been fundamental for the implementation of the ADASA 4.0 Project, which seeks the digital transformation of the institution, generating greater efficiency in Adasa's actions and better provision of service to society. In the case of SIRH, for example, we are rapidly advancing in SMART Water Management. “We want the transformation of data into useful information for management to be done quickly and securely, which will also allow us to better communicate with society and provide greater transparency in our services”, highlights the director.
EN 
PT 
