Introduction
It’s safe to say we’re not exactly huge fans of banks… as customers. Thankfully, digitalization has rescued most of us from the tedious in-person trips to the branch.Yet, for anyone who prefers the human touch, banks have ensured their branches remain open and accessible. And while standing in line at these branches might not spark joy, there’s something we do love about them: helping optimize their energy consumption.
Why Are Banks Perfect for AI-Driven Energy Optimization?
Over the years, we at Energy Twin have worked extensively with bank branches, and they present some unique characteristics when it comes to energy consumption. Why focus on banks, you ask? It all comes down to one key factor: predictability.
Unlike restaurants or cinemas, where energy use fluctuates between a full house and an empty one, banks are less affected by foot traffic. Whether there’s a queue of five or no customers at all, energy consumption remains consistent. Similarly, banks aren’t influenced by external events like stadiums, amusement parks, or concert venues, where energy usage can skyrocket during peak times.
What makes banks particularly interesting is their operating schedule. With clear opening hours, it’s possible to implement energy-saving setbacks during nights and weekends. With around 120 hours each week when branches are closed, there’s significant potential for cost savings through AI-driven solutions and insights from smart meter data.
Managing dozens or even hundreds of branches gives banks a unique advantage when it comes to energy efficiency. This portfolio scale enables centralized teams to implement impactful, long-term strategies. As organizations committed to environmental, social, and governance (ESG) principles, banks are also under pressure to ensure their energy use aligns with their public sustainability commitments.
Banks Have Their Own Specifics
Optimizing energy in banks isn’t without its quirks. For instance, popular metrics like kWh/m²/year often fall short. Here’s why:
- Unique Locations: Branches in shopping malls or historical buildings typically use more energy than standalone or modern ones.
- Size Does Matter: Smaller branches often look less efficient, with worse kWh/m²/year numbers compared to mid-sized locations.
- Special Cases: Buildings like IT hubs, headquarters, or archive rooms have energy profiles that don’t fit standard benchmarks. And sometimes, you’ll come across a branch with a truly strange energy pattern—only to learn its purpose is classified.
Rather than relying on oversimplified metrics like kWh/m²/year or weather normalization with degree days, we’ve developed machine learning-driven KPIs specifically for banks. These tools help us uncover actionable insights where traditional metrics fall short.
What Have We Found?
Our investigations have uncovered countless inefficiencies, but certain issues crop up time and again. Here are some of the most common culprits:
- Local heaters: Despite often being prohibited, portable heaters are often found in branches, sometimes left running during nights or weekends, leading to obvious unnecessary energy waste.
- HVAC schedules: HVAC systems are frequently misaligned with branch operating hours, resulting in off-hours consumption that could easily be avoided.
- Cooling setbacks: Many systems fail to implement cooling reductions over weekends, continuing to operate even when no one is around.
- Heating setbacks: Similarly, heating systems are often left running without proper off-hour adjustments, wasting energy when branches are unoccupied.
- Lighting inefficiencies: Excessive lighting is commonly left on at night, far exceeding what’s necessary for security purposes.
Beyond these recurring problems, we’ve encountered unique challenges, such as oversized uninterruptible power supplies (UPS) left behind after IT relocations, consuming energy 24/7 without serving their intended purpose.
Our AI-driven tools are specifically designed to handle these complexities. By leveraging advanced metrics and tailored KPIs, we go beyond surface-level observations to uncover deeper, hidden problems. This allows us to provide precise, actionable insights that enable banks to achieve significant and lasting energy savings across their networks.
Conclusion
While this discussion has focused on banks, the principles we’ve outlined extend seamlessly to similar institutions—post offices, insurance companies, travel agencies, and others with stable schedules and multiple locations. These organizations share a common advantage: optimizing energy consumption can be straightforward when leveraging interval data from main smart meters. With advanced AI algorithms, inefficiencies can be pinpointed and addressed without the need for additional infrastructure or disruptive changes.
Our years of experience with banks have demonstrated that effective energy optimization isn’t about sweeping transformations—it’s about precision. By leveraging data that is already being collected, we uncover hidden savings and deliver measurable results quickly. This approach integrates smoothly into existing operations, scales effortlessly to large portfolios or smaller branches, and makes even the smallest locations worthwhile targets for energy efficiency improvements. Centralized management and predictable schedules further enhance the ability to target even the smallest branches—where traditional energy-saving measures might struggle to show significant ROI.
Best of all, getting started is simple: interval data from main smart meters is all you need. With AI-driven analysis, energy optimization becomes easy, scalable, and impactful.
