Why did BAM Energy Systems decide to use UbiOps?

BAM Energy Systems has decided to use UbiOps for their data-driven analytics. “We want to use UbiOps to further develop propositions for optimizing energy consumption for our clients in both the utility and residential market.
“BAM Energy Systems uses a large variety of IoT data from devices such as solar panels, heat pumps and temperature sensors to run analyses on and optimise buildings and systems. Our data team currently uses the BAM Building Analytics (BA) platform to collect, analyse and visualise data in one central place. Since the business is expanding and even more advanced models are being developed, the need arises for an additional environment to develop data science code in a simple, scalable and cost-effective manner.

BAM uses Microsoft Azure cloud and we run UbiOps on top of that. We chose to work with UbiOps because of its simplicity and speed, and because it can be integrated with the existing analytics platform. The results are provided to Building Analytics to be visualised together with other analyses.”

The team knows that setting up a data science serving and operations backend is time-consuming. Rather than spending hours per week on managing Docker instances, cloud infrastructure and Kubernetes, they prefer to spend their time on experimenting and developing models. With UbiOps, they can deploy immediately. The Azure serving and operations backend is created and made scalable and secure with UbiOps. APIs of the models are automatically created so they can make requests to it and bring the model live without having to worry about the IT.

How does BAM Energy Systems create value for clients with Building Analytics?

Building Analytics is used to analyse and visualise all building-related data by means of automatic algorithms in one place. Until now, Building Analytics was used for analyses such as regression. With UbiOps, more advanced code, like neural networks, can now be deployed to the cloud and visualised in Building Analytics.

By combining domain knowledge and data science skills, Building Analytics finds what is relevant in the data and highlights it. This enables BAM to identify optimisations for the client such as energy reductions, forecasting demand and predictive maintenance. Consequently, BAM Energy Systems helps its clients to achieve their sustainability goals.

“Examples of data analytics are detecting inefficiencies in control strategies that would otherwise go undetected, for example, short-cycling in thousands of local climate control systems. Other analyses include forecasting energy use or (sustainable) energy production and detecting performance issues in an early stage. For example, on the Johan Cruijff ArenA, or on solar roofs. Recently we started using neural networks to forecast energy demand for our building portfolio. This is where UbiOps comes in.”

By collaborating with BAM Energy Systems we help an ambitious and motivated team to apply more advanced techniques, such as neural networks and deep learning models, that were not possible before. Moreover, we contribute to more efficient use and production of energy, thereby stimulating a more sustainable world. We hope to serve more clients like BAM Energy Systems in the coming years.