With the COVID-19 vaccines finally being distributed worldwide, travel and tourism are now poised to resurge in greater force, albeit with some interesting new changes. The post-pandemic world is one where people are now more attuned towards the future of the planet and highlighted how sustainability lies at the core of the ideal travel experience. In fact, many forward-looking hotels have already begun to shift their priorities towards developing strategies around minimising waste and resource consumption. Such efforts allow them to market to increasingly eco-conscious travellers as well as help their country as a whole to make greater strides towards achieving their net-zero emissions targets in the global fight against climate change.

As energy optimisation plays a critical role not just in greatly reducing hotel operational costs but also in achieving their sustainability goals, it comes as no surprise then that the market is flooded with energy management solution providers, all competing to prove their worth. However, throughout history, vendors and hotels alike have struggled to accurately demonstrate the effectiveness of implemented energy solutions. After all, how does one determine and measure the exact amount of an absence in cost?

In essence, it all comes down to data science. With the right data points being collected, the next step is to analyse that data and create a model that simulates how a world without the energy solution in place would look like i.e. a mathematical model that accurately predicts how much energy would be used by the hotel if the vendor’s energy solution was not implemented. From there, energy savings over a specified time period can then be calculated by looking at the difference between the model’s prediction and the hotel’s actual recorded energy consumption for that time period (i.e. via energy meters installed, or by looking at hotel energy bills).

Fortunately, with the advent of IoT (Internet of Things) technology, it is now possible to connect multiple measurement devices and collect an immense amount of interconnected data of any system such as hotel guest rooms environments and hotel HVAC energy systems. In addition, advanced Artificial Intelligence software can then be used to process and analyse enormous amounts of data collected to greatly tweak and refine predictive models that accurately simulate how changes in the environment (e.g. the installation OR absence of an energy solution) changes the way a system works (e.g. how much energy a hotel room’s HVAC system consumes).

So how have hotels been evaluating their energy management programs and tracking sustainability efforts so far? It turns out, there are two key methods currently being used in the industry when it comes to calculating energy savings in hotels.

Comparing Two Key Methods of Calculating Energy Savings

In a simplistic world, hoteliers might just compare monthly electricity bills to know whether energy-efficient HVAC systems and other green solutions are saving them money. Unfortunately, it is not this simple. The energy consumption of a hotel’s HVAC system(s) can be greatly affected by humidity levels, guest setpoint temperatures, outdoor weather, occupancy levels and a variety of other critical factors.

Even with smart hotel technology, is it reasonable to do “before vs after” comparisons without considering occupancy levels or even guest temperature preferences? These factors and many others can greatly dictate how much energy an HVAC system will need to cool or heat a room. So in order to measure the actual reduction in energy consumption by a solution provider, these additional factors and how they impact energy use must first be understood.

The International Performance Measurement Verification Protocol (IPMVP) – which is widely known as the industry gold standard to follow – suggests two standard approaches for solutions providers to calculate and estimate energy savings. A brief summary comparison of the two approaches is below:

The Traditional Method: Historical Baseline Approach

The most commonly used method of measuring energy savings is the Historical Baseline Approach. This method creates a simple straightforward baseline model to predict how much energy a hotel will typically use in the current year by first looking at the pattern of their past years’ electricity bills. If historical weather data and/or monthly occupancy rates are also available, then these might be included to improve the model’s ability to predict similar patterns of energy use for the current or future years to serve as a baseline.

With this baseline model, solution providers will then predict how much energy a hotel would use in the current year had the hotel not installed their energy management solution. Thus, if a hotel’s energy bills for the current year are lower than the predicted baseline, then the solution provider would claim the difference as the energy savings achieved.

Pros of Historical Baseline Approach

The biggest advantage of the historical comparison approach is that it offers a relatively quick and easy way to measure energy savings. Even if a solution provider goes the extra mile of including average occupancy rates to adjust their predictive baseline model, it’s not difficult to compare yearly costs prior to a solution’s implementation with a straightforward model’s estimate of yearly costs afterwards.

Cons of Historical Baseline Approach

Unfortunately, there are several issues that can make the Historical Baseline Approach less reliable than other approaches. Because of these drawbacks, it’s possible to believe your hotel’s smart energy technology is working when in reality nothing much has changed. Here are the biggest problems with this approach:

  • Reliable and fine-grained historical data can be difficult to come by, making it difficult to accurately estimate real improvements in energy efficiency based on past electricity bills and weather data alone.

  • Even if one takes into account historical occupancy rates to predict a more accurate baseline estimate, this still leaves out other important factors such as humidity, guest preferences and ambient room temperature all of which can greatly affect energy equipment efficiency.

  • Historical energy bills also tend to be reported for the whole building – they rarely provide room-by-room data. This means you can’t limit the baseline comparison to just the energy use differences achieved in the affected areas of the hotel property alone.

  • Using sparse and limited historical data typically has a much lower level of accuracy – around 70 per cent – with error ranges up to +/-15 per cent.

The Advanced Method: Simulated Baseline Approach

With the evolution and widespread use of IoT devices, the technology needed to advance energy calculation methods is now at hand. The smart hotels that have the most success with verified energy savings are those that have a wealth of relevant data from their property to enable the use of the Simulated Baseline Approach. This is the more advanced of the two methods suggested by the IPMVP. This baseline model directly measures several key environmental variables which are then used to simulate energy consumption patterns that are unique to each hotel’s rooms.

Unfortunately, the Simulated Baseline Approach is still rarely seen in the industry due to its complexity to execute. Many energy management solution providers exclusively use the traditional approach because it’s logistically much easier and far less costly to create a baseline model using existing historical data. In contrast, the Simulated Baseline Approach often requires a provider to install expensive equipment to continuously collect a multitude of new data variables to build an up-to-date model that closely predicts the true baseline energy consumption of a hotel’s rooms so that it can then be compared with a hotel’s current room energy usage.

Pros of Simulated Baseline Approach

  • Identifies and takes into account many key factors that affect HVAC energy consumption when creating a predictive baseline model.

  • Involves continuous measurement of current HVAC energy consumption and other key factors which are used to simulate baseline usage.

  • Predictions are specific to guest room HVAC usage so unrelated energy use from other parts of the hotel not controlled by the solution provider will not affect the energy savings estimate.

  • Fine-grained hourly data can even be collected for a more accurate simulation of baseline energy use patterns.

  • Has up to a 95 per cent level of accuracy.

Cons of Simulated Baseline Approach

Unfortunately, the complexity and cost of collecting and analysing vast amounts of data mean most solution providers are not able to use the simulated baseline approach. With the traditional Historical Baseline method, though, the energy efficiency estimates of your smart hotel could be grossly inaccurate. However, the simplicity of the Historical Baseline approach sometimes appeals better to hoteliers as well as key hotel stakeholders. This is because they may find it more difficult to fully understand and verify the complex Simulated Baseline Model(s) developed by the solution provider.

The Simulated Baseline Model is the Future of Savings Calculation (and SensorFlow is a Specialist)

There’s no denying that traditional methods of measuring energy savings just aren’t as reliable as the Simulated Baseline Approach. This is exactly why SensorFlow utilizes this method by creating a predictive baseline model that takes into account all critical factors that affect guest room HVAC systems to simulate an accurate baseline of energy consumption for our hotel clients.

There are many reasons why smart hotels around the globe have trusted SensorFlow to identify their HVAC energy savings:

  • All critical contributing factors are continuously measured (e.g. guest A/C setpoint, humidity levels, room occupancy, external weather, internal room equipment efficiency and more).

  • Our family of sensors track all critical energy and environmental data variables for every guest room at 15-minute intervals all day, every day.

  • Our predictive model takes into account over 200,000 monthly data points collected on average per hotel.

  • Data points identify energy consumption on a per-room basis, so it prevents miscalculations from unrelated energy use while also allowing smart hotels to identify problems in individual suites.

  • SensorFlow software uses artificial intelligence and machine learning to drastically improve our predictive baseline model’s accuracy.

  • A unique prediction model is custom generated for each hotel, and we always test it to ensure it reaches the 95% accuracy mark.

SensorFlow ensures that when you’re looking at energy savings after implementing our solution, you’re only seeing real savings directly stemming from our automation solution. In fact, an independent review by the EarthCheck Research Institute has verified our savings calculation method to be both valid and accurate. This is why all our clients know and trust that our savings calculations are fair, transparent and reliable. Watch this short video to learn more about our savings calculation methodology or check out our infographic to find out how SensorFlow can help you be a smarter, greener hotel.

If you want a more in-depth technical analysis on how SensorFlow generates reliable savings estimates for your smart hotel, you can read the entire report on our calculation methodology here. If you’re ready to implement smart energy-saving strategies and be a smarter, greener hotel, simply reach out to us today at [email protected].