The Tally Digital team was approached by Simply Energy Australia with a request to provide a solution that empowers customers to manage their energy usage effectively and optimize power consumption. During the collaboration, load disaggregation emerged as a promising option. This innovative approach offers valuable insights into energy consumption patterns and opens up possibilities for significant cost savings. The best part is that it does not require any additional smart-home appliances, making it a convenient and accessible solution for customers.
Our data scientists conducted extensive research on smart meter technology, energy consumption patterns, and customer needs. We thoroughly analyzed industry reports, as well as insights from Simply, to gain a comprehensive understanding of the evolving energy landscape.
During our research, we recognized that accurate and valuable insights could be achieved by leveraging artificial intelligence (AI), machine learning (ML), and deep neural networks (DNNs) to break down energy consumption to the appliance or category level. To enhance the user experience, we also incorporated a simple infographic to effectively inform the customer.
To put our findings into practice, we designed a pilot user experience involving a sample of Simply’s customers. Our main objective was to provide actionable insights that would enable customers to make informed decisions and optimize their power usage. To ensure the success of the pilot program, we conducted real-world monitoring, gathered user feedback through a survey, and tested the model for accuracy. This approach allowed us to refine our load disaggregation method, ensuring both data accuracy and an improved user experience.
Our research confirmed a promising opportunity to create an engaging experience for customers. This involved presenting them with detailed insights, empowering them to understand the cause and effect of their actions around energy usage.
Collaborating with Simply Energy Australia, the Tally Digital team developed Tally Load Disaggregation – an innovative solution utilizing AI, ML, and DNNs to detect appliance consumption. Our solution consists of three major components: a deep neural network for low-medium consumption appliances, long-short term memory (LSTM) and convolutional neural networks (CNNs) for electric hot-water systems and timed-loads, and a combination of classifiers and regressors to detect and quantify appliance consumption. This data was presented as a bubble chart, in-line with Simply Energy’s friendly branding and tone of voice.
Tally Load Disaggregation delivered impactful results:
Empowered Customers: Clear graphs and visualizations enabled customers to pinpoint factors contributing to high power bills. This empowered targeted actions for energy optimization.
Faulty Appliance Detection: Customers could identify faulty or inefficient appliances, making informed decisions about repairs or upgrades, potentially leading to energy and cost savings.
Scalable Implementation: Tally Load Disaggregation seamlessly integrated with the existing Simply Tracker product, providing granular usage data for monitoring by hour, day, and week.
Our collaboration with Simply Energy Australia led to the successful development and deployment of Tally Load Disaggregation. By leveraging AI, ML, and DNNs, customers gained control over power consumption and cost savings. The solution provided actionable insights, boosting satisfaction, loyalty, and reducing customer churn.