Comgás modernizes its technology park to operate 100% in the cloud and expands the use of advanced data analytics

With a history dating back to 1872, Companhia de Gás de São Paulo (Comgás) is the largest natural gas distributor in Brazil, delivering canalized gas to the residential, commercial, industrial, automotive, cogeneration, and thermo generation segments. The company serves more than 1.7 million customers in 88 cities, passing through the metropolitan region of São Paulo, Baixada Santista, and Vale do Paraíba—in addition to the administrative region of Campinas. 

Comgás went through the process of renovating its technology park and decided to modernize its architecture, taking its infrastructure to the cloud, moving away from on-premise servers. At the same time, it also built a robust environment to better work with data generated in large volumes in the company, to assist decision making and, consequently, improve delivery to its customers. 

Lucio Oliveira, CIO of Comgás, celebrates the partnership and states that the Microsoft Cloud was crucial to modernizing Comgás’s technological framework and aims to deepen its level of digital transformation. "Our goal was for technology to play a leading role in all the company's activities, and that's what we've done with this project,” he says. 

Efficiency, governance, speed: Azure empowers the data and AI team 

The first step was the creation of a data environment supported by Microsoft and partner companies, which assisted in project planning and training the Comgás team. According to Thiago Rolemberg, head of the company’s Data and AI Department, the initiative was intended to improve the volume of data, which could strengthen the company’s overall operation. “We realized the possibility of extracting data better and more efficiently using AI,” he explained. “And, thus, better understand our customers to make more assertive decisions.” 

Comgás, together with Microsoft, has implemented a modern analytical platform in Microsoft Azure, including cloud services that offer agility, flexibility, and governance for the management and analysis of the company's data. “We are constantly looking for efficiency, and having a robust partner like Microsoft who knows and understands our operation helps us to better organize our architecture,” Rolemberg adds. 

Since the implementation, more than 55 TB of data have been processed, and the platform gathers 60% of the representative data of Comgás. Approximately 30% of the company is already using the environment, and the number is expected to grow to 55% by the end of the year. “As we have a robust and reliable platform, we have started developing AI algorithms and gained in efficiency, data governance, and volume,” celebrates Rolemberg. 

Another benefit was Power BI, which homogenized the presentation of data disseminated by the company's teams, which work only with approved data reports. Finally, the data extraction process that originally took between 8 and 15 days was reduced and can be done in real time, depending on the area involved. 

Datacenter to cloud migration: A modern, robust, and scalable environment 

Marcos Vinicius Cavalcante, head of Corporate Architecture and Comgás Cloud Infrastructure, says that the data environment was the company’s first step towards cloud architecture, followed by the migration of the data center in record time—six months.  

“Our data platform was already being built, but modernizing the entire environment was not a reality,” explained Cavalcante. “That’s when we outlined a specific operating model.” Comgás opted for a solution in a Microsoft environment due to the proximity of the companies, mutual knowledge between the organizations, the familiarity of its team with the tools, and the soundness of the platform. “It’s a much more modern, robust, and scalable environment that doesn’t limit us to having a usage ceiling or condition that disrupts our business due to upgrades,” he said. 

The focus of the implementation was the search for a new operating model that gave the company new possibilities. They gained access to tools like Azure Machine Learning, which accelerates and manages machine learning cycles in the cloud, and Azure DevOps with developer services, where teams can plan work, collaborate on code development, and deploy applications. 

“We’ve come to the cloud environment with a range of product and service options that we don’t buy, but rather work through usage and consumption plans—for the same or less value,” explained Cavalcante. “Our processing capacity and availability has become scalable and more efficient as we no longer have idle technology resources.”  

A technological transformation seeking efficiency and focused on people 

In terms of cost, Oliveira assesses that the cloud alleviates the need for renewal of the technological park—which in itself would have a high investment—and would still have an expiration date, with the need for a team of professionals responsible for periodic (and also costly) maintenance. “With the cloud, I have an always-updated architecture, and the security capability to protect my infrastructure is also high,” Oliveira explained. 

Oliveira also said that the company recently underwent a culture change, which went from the customer at the center to the concept of people at the center. Thus, the entire journey of digital transformation and the search for efficiency is driven by the same purpose, just like in the case of the cloud and a more sophisticated data analysis with AI and machine learning. “Efficiency is a great benefit,” said Oliviera. “The model with learning capacity is improving all the time, by its design nature it will propose new adjustments, parameters, and information to make decisions. It’s a continuous improvement job.” 

Among the next steps of this digital transformation, the data area aims to bring more tools to further foster the company’s data-driven culture, while the infrastructure area is evaluating other systems, some of Comgás’s core, to bring everything to the cloud. “We rely on Microsoft at all times to help us with this training—with continuous training and the training of the team that has been happening very satisfactorily,” Oliveira concluded.