from Steve Baab, DNV GL
Can data analytics replace people in energy efficiency?
Big data and analytics have been buzzwords for quite some time now. Within the utility space, Advanced Metering Infrastructure (AMI) has opened up new opportunities to collect and use energy usage data. In the past, most utilities only collected usage information based on the type of meter and the reading schedule. Different meter types collected different amounts of information. Larger customers would have energy usage (kWh) and demand (kW) in various intervals (monthly, hourly, every 30 min…). For all customers, the reading schedule was usually monthly and was performed manually by a meter reader. With AMI systems there is an opportunity to collect data without sending someone out into the field, and to collect the data whenever a utility wants it, rather than on a prescribed schedule. Utilities implementing AMI systems usually install three major components: the meters, the network to communicate with the meters, and the meter data information system. For a utility to decide to install an AMI system, the benefits (labor savings, operational efficiencies, customer satisfaction) need to justify the costs.
One of the largest benefits of installing AMI networks on the utility side are labor savings. Utility meter reader jobs are perhaps the most significantly impacted with an AMI network, but there are other areas of labor savings. The ability to pinpoint outages, reduce bill estimation work, and remotely disconnect service are also areas where an AMI network can reduce labor costs. For customer satisfaction, having interval data available allows a utility to offer its customers more flexible billing options and enhanced information on consumption.
But what about putting the interval usage data to work for energy efficiency? Can the increased granularity of usage data save labor within efficiency programs? There’s a decent amount of information housed within all that data. When interval data is combined with external data such as building type, size, and age, an analyst or maybe even simple computer algorithms could provide savings recommendations. This could potentially save labor by not having to visit customers to find savings through an audit or assessment.
Although analytics is extremely helpful in identifying for potential customers where there may be energy efficiency opportunities, I don’t think it will ever evolve to the point where it will replace people (or at least within the next 10-20 years). Here’s why:
Existing load information can’t determine exactly what’s creating the load pattern.
DNV GL and others are getting good at identifying load shape patterns using advanced data. Detailed load shapes support impact evaluations, potential studies, and informs efficiency and demand response programs. However, on a customer by customer basis, analytics can only provide directional vs. actionable information. Analytics can only point you in the right direction, but you’ll have no way of knowing for sure without a visual confirmation.
Inability to determine the exact type of equipment generating the loads.
Software and analytics are getting very sophisticated at identifying load patterns, but there is still uncertainty in customer behavior and equipment that creates the load pattern. An inefficient cooling system or an occupant’s preference to have their building two degrees cooler than everyone else can create similar load patterns. (Also poor insulation, high internal loads, or anything else driving abnormal cooling loads.) Analytics can provide good directional indicators based on interval meter data, but only a person knowledgeable in energy efficiency and the site’s equipment can provide tangible recommendations to reduce energy use.
Human interaction drives interest in energy efficiency
Reports, emails, text messages, or web portals can all help drive some interest and awareness, but nothing triggers a call to action like a real person. For the same reason sales jobs are not going away anytime soon, the role of energy efficiency experts who perform audits and/or assessments will also not go away in the near term. During an energy assessment, the customer and the energy efficiency expert will get a chance to interact. The energy efficiency expert will learn about the customer, the customer will learn about energy efficiency, and together they will establish mutual trust and then develop the best course of action to save energy.
In summary, analytics may be able to uncover some of the technical potential of energy efficiency and guide staff on which customers to visit, but it will never be able to gauge the interest and motivation of customers that have the choice to spend time and money on energy efficiency (or not). Getting people to become more energy efficient requires a conversation with the people who are using energy. There is a big difference between buildings that have the technical potential to save energy and customers that have the interest and act upon recommendations. In my experience, face to face conversations work much better than computer algorithms.
Steve Baab: Steve works at DNV GL as the Service Line Leader for Program Development and Implementation. Steve works to bring best practices and innovative program ideas to DNV GL clients. Before coming to DNV GL, Steve was the ComEd C&I portfolio manager where he led the Smart Ideas program from initial launch in 2008 to a program that offers a wide spectrum of customer solutions in both energy systems and market segments.
Originally posted here
About Jürgen Ritzek
Juergen Ritzek is co-founder and Business Director of EEIP. Juergen is responsible for strategy, marketing and business development of EEIP and drives the growth of EEIP towards an energy transition platform. Juergen leads EEIPs B2B communication and relations and ensures EEIP relevance for value chain challenges (inter-company) and internal decision-making processes (intra-company). Following an international career at Unilever he founded European network consultancy GBC (2009) and EEIP (2011).