In the previous post we have established that cutting energy cost in quick-service restaurant is well-worth owner’s attention. We have also established that about 80% of electricity and 100% of natural gas are consumed by refrigeration, cooking, HVAC, and water heating, with refrigeration and cooking leading in both categories. These are all areas where hard-to-notice malfunction or operational error may easily cause an energy overuse, aka waste.
Finding such issues is a routine maintenance task … if one knows where, when and what to look for.
Looking for all operational issues everywhere all the time is prohibitively expensive … unless you are a tireless algorithm!
Here is when AI algorithms come to the rescue: while for a human it is boring to monitor every fridge and every thermostat, it is a routine task for an AI algorithm to find wasteful restaurants by comparing consumption of restaurants in a big pool.
If 500 QSRs serve the same menu using similar equipment, their energy consumption after weather adjustments must be similar too. If some QSRs consistently use more energy – it’s time to inspect their equipment condition and operational procedures.
AI algorithms can go further and provide a preliminary diagnostic of energy waste root causes. Here are several oversimplified examples:
- Higher electric load 24×7 – fridge gasket is damaged, or evaporator is dirty, or ventilation runs non-stop
- Load bumps at night – faulty controls or wrong operational procedures for cleaners
- Higher load in the morning than in the afternoon – shifts operate differently or higher level of operation
Algorithm or not, just ‘show me the money’
Use of AI is popular nowadays for a good practical reason: it is a cheap and effective way to find energy waste in big portfolios of similar establishments.
An AI-based analysis of electricity and natural gas consumption data performed by kWIQly GmbH (Switzerland) at 900 pubs in the UK has revealed that 44% of pubs used 30-50% more energy compared to their peers. Average excess monthly cost amounted to 1,625 pounds or nearly half of what a typical pub owner makes.
Most of this waste was operational – settings, leaks, gaskets, dust on filters, dirt on coils, etc. Once the analytical part was done – problematic pubs were found, price tags were attached to waste and likely causes of waste identified – fixing issues became a routine maintenance or management task.
Now the fun part.
Monthly investment into consumption data analysis in the pub engagement was about the same as a price of a pint. How many pints (burgers, muffins) should a pub or a quick service restaurant sell to put 1,625 pounds into owner’s pocket?
Is investment into energy use monitoring worth it? – Do your own math.