Better energy efficiency policy with digital tools

Energy efficiency requires policy makers to interact with a diverse set of different stakeholders including end users, businesses, utilities, information technology (IT) companies, energy service companies (ESCOs) and data providers. Developing policies that are both broad enough to effect change on a large scale and targeted to meet the needs of such diverse groups requires detailed data and a level of connectivity which is difficult and costly to achieve. Digital tools can be used to provide easier access to such data and to foster connections necessary for the next generation of energy efficiency policy.

Potential exists to leverage digital tools at all stages of the policy cycle of design, implementation and monitoring. Their growing role is set to enable policy makers to measure the value of integrated energy efficiency and distributed energy resources more transparently, to expand the energy efficiency ecosystem, and to allow new innovative market-based policy approaches.

At the policy design stage, digital tools can provide access to more granular and real-time data, and advanced analytics and modelling capabilities can help predict the impact and cost-effectiveness of programmes. During programme implementation, digitalisation can be an effective communication tool to enable more user-centred policies. At the same time, digital tools allow for data to be produced at a much higher frequency and larger scale than before, allowing for a more continuous approach to evaluation.

This article highlights examples from the following ten broad groupings of digital technologies and tools, exploring how they are being used in different contexts around the world:

• databases and big data
• non‑intrusive load monitoring
• GIS mapping and remote sensing
• virtual buildings and digital twin cities
• virtual audits
• digital certification and compliance
• digital communication and networking
• smartphones and apps
• natural language processing
• web search analytics.

Databases and big data

Huge quantities of energy data are being generated as the world digitalises through the roll-out of “smart” devices, grids, and other IT infrastructure and systems. Such data can provide a valuable resource for policy makers looking to deliver on their energy and climate policy goals.

For example, the government of India has developed India Energy Dashboards (IED), an open-source portal that collects data and monthly reports on the country’s electricity, petroleum and natural gas use. India’s Building Energy Efficiency Program Dashboard is another digital tool which provides policy makers and others with transparent and accessible data, increasing understanding of the energy system and supporting energy policy decisions at the highest level. The country’s government also created the National Ujala Dashboard with the objective to encourage energy-efficient devices in residential facilities by raising awareness among consumers. The online database displays the number of LED lights installed by region in real time, along with the respective annual cost savings, annual CO₂ reductions and avoided peak power demand.

Since 2013, the Chinese government has been prioritising online energy monitoring systems as part of its Implementation Plan of the Top-10 000 Enterprises Energy Conservation and Low Carbon Program and through the Notice about Online Energy Monitoring Pilots in Key Energy Using Entities. Public-sector buildings across all levels of government have been implementing such systems to provide real-time energy data, allowing for the automation of energy management and a transparent method for monitoring energy efficiency actions. Most online energy monitoring systems for the Chinese public sector are currently individual systems. The next stage of policy evolution will be to move to an approach that provides a more integrated system-wide view.

For example, Hangzhou’s local government, together with Alibaba Group, has implemented the City Brain project to improve transportation energy efficiency. Under the project, a cloud platform captures images from connected street cameras, translates them into traffic data, analyses the results and comes up with the most efficient solutions via algorithms. This information is then used to achieve optimal efficiency through smart traffic lights. The implementation of the City Brain project has reduced congestion in Hangzhou by 10%.

CalTRACK in the United States is an example of an analytical platform supporting the energy efficiency market and other demand-side resources. CalTRACK uses open-source software and empirically grounded analytical techniques in conjunction with meter-based changes in energy consumption data to calculate avoided energy use. This can be used to support support energy saving and CO₂ mitigation policies by better enabling the procurement of energy efficiency, electrification and other distributed resources. Such platforms have supported the California Public Utilities Commission’s adoption of a system replacing “energy savings” with “total system benefit” as the metric for energy efficiency goals. A market for such values can be enabled through digital tools such as Recurve’s Engine, which can be used to quantify different elements including energy and CO₂ savings, technology support, and broader social goals targeting vulnerable communities.

A core feature of this approach is the open source nature of methodologies and software, which provides a transparent and robust method for the calculation of both efficiency programmes’ impacts and measures required to bring energy efficiency and demand flexibility to the market. For example, the Linux Foundation’s LFEnergy OpenEEmeter project calculates normalised metered energy consumption and demand savings. Results can be accessed and adapted as needed by jurisdictions at very low costs compared with more traditional approaches. Moving towards common metrics for calculating the benefits of energy efficiency can also help scale up markets and financing for system flexibility resources, such as virtual power plants and Pay-for-Performance (P4P) approaches to energy efficiency resource acquisition.

The SENSEI project is the first in Europe to adopt a P4P financing scheme. The model allows utility operators to repay building energy efficiency investments based on actual energy savings through the installation of smart meters. The P4P concept not only incentivises private financing into green and efficiency-enhancing projects, but also brings added value to energy retrofits.

In the United Kingdom (UK), the Green Finance Institute is exploring schemes to leverage smart meter data to establish a standardised, industry-recognised protocol to measure energy savings in UK buildings. Such schemes can support greater use of market-based energy efficiency policies by enabling new forms of contracting and procurement among energy utilities, networks and aggregators. They can also support government retrofit policy goals by helping scale up efficiency investments through new financial products and business models.

In 2020, Australia launched its Distributed Energy Resource Register, which provides a database of information about distributed energy resources (DER) devices installed in the National Electricity Market. This is foundational information for the Australian Energy Market Operator’s (AEMO’s) DER Program. The visibility of installed DER devices allows AEMO to better manage the electricity grid and ensure reliable, secure and affordable energy for all customers. Supporting this work, Microsoft has provided a cloud-based platform for big data collection and analysis to allow Australia’s energy market to transition from 30-minute to 5-minute settlements, removing several market barriers for renewable energy and demand-side resources. The new technology will also allow AEMO to process more than 7 million meter reads per day in around 45 minutes, a process which currently takes several hours.

Non-intrusive load monitoring

In India, policy makers at the Bureau of Energy Efficiency use a digital data collection and analytical tool called the National Energy End-Use Monitoring dashboard to obtain up-to-date information on appliance energy use. Real-time metered data from 200 households is combined with survey data conducted in 5 000 households to provide policy makers with a detailed understanding of which appliances are being used where, and what impact their use is having on energy demand.

In 2018, the government of the United Kingdom (UK) invested around USD 6 million in the Smart Meter Enabled Thermal Efficiency Ratings (SMETER) Innovation Programme. SMETER aims at developing, monitoring and evaluating smart meter technologies applied to households’ appliances. Participants are guaranteed complete anonymity, and results will contribute to both assess the technology’s performance and raise awareness among residents.

GIS mapping and remote sensing

The buildings sector contains significant potential to improve energy efficiency. For example, in the IEA Sustainable Development Scenario, the energy used per square metre of building floor area decreases globally by at least 2.5% per year on average. This can be achieved as a result of more efficient new buildings, deep retrofits of existing buildings, a tripling of heat pump uptake and a 50% improvement in average seasonal performance of air conditioners as well as other energy efficiency measures. However, identifying exactly which buildings would benefit from government policy intervention has generally required conducting costly and detailed analyses of a city’s building stock, often including in-person audits. Thanks to digital technologies, identifying energy efficiency potential in buildings is increasingly possible at much lower cost.

These digital tools have the potential of making it applicable on a larger scale for large roll-out of efficient housing and other building construction. In Europe, pilots are being run to use geographic information system (GIS) mapping software to identify high-priority areas for efficiency upgrades to heating and cooling systems through the Hot Maps project. The open-source tool allows city planners to visualise geographical areas with potentially high heating or cooling loads, which could then be prioritised for energy efficiency upgrades as part of heating or cooling action plans. While currently being piloted at the city level, the tool has the potential to be scaled for regional- and national-level policy makers. The project was followed by Decarb City Pipes 2050, an initiative to create roadmaps to decarbonise urban heating and cooling.

In 2017, the Indonesian Ministry of Energy launched the ESDM One Map online application, which displays thematic maps of the country’s energy sector. The web GIS maps support policy makers in assessing the scope for energy efficiency investments by covering topics such as coal, oil and gas, as well as renewable energy, electricity networks and environmental disaster risk areas.

Digital tools can also be used to combine administrative data collected from government policies and programmes to determine buildings’ adherence to minimum performance standards. The city of London developed the London Building Stock Model, which combines energy data with data on building size, activities undertaken within them, their construction materials and their energy systems. The model will be extended to the whole United Kingdom, to determine which buildings comply with building regulations, identifying where energy efficiency interventions are most needed.

In North America, MyHEAT is using remote sensing data, combined with advanced algorithms, to identify buildings where heat losses are prevalent. Data from high-resolution thermal imaging sensors taken from aircraft are corrected for local changes in temperature, microclimate and elevation, and processed by machine-learning algorithms to show where energy is escaping from all buildings in a city, to create a comparative model.

The Canadian platform also engages with users to direct them towards energy efficiency incentive programmes, rebate information, online marketplaces and energy audit resources. During the Medicine Hat project in 2018, residents were randomised into treatment groups, each receiving different feedback on their house energy use. The treatment group that was shown heat loss maps and comparison ratings experienced 30% higher participation in weatherisation programmes, and 19% more online rebates for other home upgrades compared with the control group. The high-heat-loss members of the MyHEAT treatment group also reduced their energy consumption by 4.4% compared with the control group.

By combining such data with socio‑economic information about householders, policy makers can better target energy efficiency retrofit programmes to the buildings and people that need them, faster and more cost-effectively than conducting in-person audits. By doing this at the policy design stage, retrofit policies could be better designed, reducing the risk of government over- or under-spends. The ability to target specific homeowners (for example low-income homeowners with high heat loss) helps policy makers focus on the homeowners that need the assistance most, and reduces participation by free-rider households that can otherwise afford to do the retrofits.

MyHEAT also works with utilities looking for non‑pipeline (NPAs) or non‑wires alternatives (NWAs). Utilities use the MyHEAT data to identify and target inefficient homes in areas of their systems that have peak day load restrictions, and in areas where the utility is looking to expand its offering. NPAs and NWAs can help eliminate the need for large capital projects such as new pipelines, transmission lines or substations.

Satellite data combined with machine learning and building simulation tools have also been used to inform India’s Cooling Action Plan and the assessment of the proposed smart city of Amaravati. This has enabled analysis of possible urbanisation pattern and associated heat island effects.

Virtual buildings and digital twin cities

At the city level, so-called “digital twins” (virtual models of a city’s systems) allow policy makers to test how planning decisions may affect a city’s infrastructure, people and resource use, including energy efficiency.

This can be done for an individual system (such as an energy or transport system), the entire city, or potentially an entire national building stock. For example, planners in the city of Versailles used a digital twin of the city’s road network to model the city’s freight logistics flows and formulate a plan that would reduce congestion for delivery companies, improve energy efficiency and reduce air pollution.

Existing delivery zones were partnered with mobility data on the actual traffic flows and delivery routes in the city (including third-party smartphone data from trip-planning apps). Planners were then able to simulate how changing the location of delivery zones might affect congestion and local air quality. The city government estimates that using digital tools reduced the normal time for conducting such analysis by 80% while the changes are estimated to result in a 10-30% reduction in congestion related to double parking.

The local government of São Paulo, Brazil, also initiated a digital twin project to create a virtual replica of the city rural area’s electricity grid. Over 5 000 sensors were installed on the network, communicating its status in real time, helping predict faults and guiding grid investments and maintenance measures.

At a larger scale, the government of Singapore has recently completed a full-scale digital model of the entire city, Virtual Singapore, including 3D digital replicas of every building in the city. For city planners focused on energy efficiency, the digital twin provides the capability to accurately simulate how new developments and planning changes in the city might affect a range of energy-related indicators, including solar irradiance, road and foot traffic flows, heating and cooling needs, and many other factors. Given the city’s size constraints, the digital twin provides an extremely useful alternative to testing planning interventions in the real world.

Among many other digital technologies applications explored by Natural Resources Canada (NRCan), the NRCan ecoENERGY modelling software was used to design the currently most energy-efficient office building in the country. The virtual version helped visualise the long-term energy and economic benefits of efficient systems. The building is made of recycled, locally sourced materials and leverages smart sensors to achieve an optimal energy performance. The design includes heat-recovery systems, which reuse waste heat from the server operations room through heat pumps.

Virtual audits

As a response to the Covid-19 pandemic, virtual building code inspections have become a useful tool to safely conduct compliance activity. Such inspections also offer the potential to provide expertise in remote areas and in areas where capacity to conduct such audits may be constrained. As they are also cheaper to perform, it is likely that they will continue to provide value beyond the restrictions imposed by the pandemic.

Remote inspections make use of a broad range of digital tools: from Internet of Things (IoT) sensors to monitor and prevent maintenance, to digital twin and 3D modelling. For example, Dubai will continue to conduct virtual inspections even after the pandemic, helping promote cost and energy saving.

Italy’s National Agency for New Technologies, Energy and Sustainable Economic Development released two smartphone and tablet applications that conduct building energy efficiency analysis in schools and residential buildings. The applications are currently accessible only to technical experts and provide accurate measurements that compare the structure in question with national standards. Then, users receive a list of concrete suggestions to optimise the building’s energy performance.

Digital certification and compliance

For energy efficiency standards and labelling programmes, digital technologies provide the possibility of “electronic” labels, which solve the problem of having to physically update labels on appliances and equipment when standards are updated periodically. QR codes coupled with smartphones and apps can provide a more effective certification system compared with traditional product labels. With QR codes attached on appliances and linked to a database, consumers can easily check and compare the energy efficiency of appliances. Appliance labels in both the European Union and the People's Republic of China include QR codes that consumers can scan with their smartphones to easily obtain more information than is presented on physical labels, including product comparisons. Manufacturers benefit by being able to update product information and receive consumer feedback, market surveillance and supervision officials can gain easier access to the product registration database, and policy makers can more accurately and efficiently monitor product quality. India also operates an online product registration system which allows the fast and transparent assessment of equipment in the market.

India also uses the web-based PATNet portal to support its Perform Achieve Trade market-based mechanism to reduce specific energy consumption in energy-intensive industries. PATNet allows large energy users in industry to complete their energy return forms online instead of sending them via email or by hard copy. This portal also tracks the trading activity of energy saving certificates for each registered participant.

In South Australia, a digital certification of compliance system for electricians, the eCoC (for electronic certificate of compliance), recently replaced the well-established paper certificate regime. An 18‑month transition period was allowed to move to the new more cost-effective and reliable digital system. This system is also integrated to support data collection for the new national DER Register. This approach minimised duplication of data entry by electricians, and provided a stream of highly granular data around DER installations to the Office of the Technical Regulator within the Department for Energy and Mining. For example, in the first week of June 2021, almost 600 eCoCs were submitted for work on DER. This provides some useful insights into the quantum of installs, location and aggregate generation capacity. Perhaps the greatest use of this data, and of eCoC data in general, has been to inform and enable compliance activities such as desktop audits of inverters and sites where solar generators were installed without seeking authorisation.

Digital communication and networking

Digital communication and engagement tools allow energy efficiency policy makers and regulators to provide up-to-date information more effectively on the energy efficiency performance of appliances, vehicles and buildings.

The UK government launched Simple Energy Advice, an online platform advising citizens on how to reduce their energy consumption and electricity bills, guiding them towards optimal energy efficiency grants. All recommendations are based on the country’s standards on houses and energy use.

The South African online campaign Smart Buildings also aims at connecting stakeholders to deliver green and user-centred regulations on smart buildings. The digital hub provides tips and guidelines on energy efficiency, water and waste management, indoor air quality, and sustainable materials that citizens, professionals and governmental entities can consult easily and for free.

Providing tools for viewing and exploring datasets can also help make data more accessible to third parties, potentially encouraging private energy efficiency investments and increasing transparency. A good example of this is the Building Performance Database in the United States (US), an online tool that the US Department of Energy, in collaboration with the Lawrence Berkeley National Laboratory, created to help people access and browse data on building energy performance, from governments, utilities, energy efficiency programmes, building owners and private companies. The website allows users to explore the data across real estate sectors and regions, compare various physical and operational characteristics, and gain a better understanding of market conditions and trends in energy performance.

The US Department of Energy also offers the Database of State Incentives for Renewables and Efficiency (DSIRE) to collect incentives and policies on renewables and energy efficiency by state. DSIRE aims at providing professional financial guidance on investments and tax decisions.

Similarly, the European Commission funded the De-risking Energy Efficiency Project, an open-data platform designed to provide the data necessary to increase energy efficiency investment from businesses and individuals, helping to link investors with bankable energy efficiency projects. Data providers are a combination of companies, think tanks and governments. As the number of data providers increases, the tool will help developers and financiers to assess the monetary risks and benefits of energy efficiency business opportunities.

Due to the Covid-19 pandemic, the 30th annual Chinese Energy Conservation Week was held online in 2020. Between 29 June and 5 July 2020, the Chinese government enlisted e‑commerce platforms to promote energy-efficient and environment-friendly appliances. Top appliance suppliers, including Gree and Haier, published videos to introduce their energy-efficient products and efforts in energy efficiency innovations. E‑commerce platforms such as Tmall, JD and Suning announced discounts and other incentives for purchasing appliances with high energy efficiency ratings. The Chinese government has also published comics and emojis on energy-efficient choices via mobile apps and initiated a knowledge contest for smartphone users with rewards.

Smartphones and apps

The Greater London Authority has set up a portal which combines data collected from its own bus and rail networks with mobility data from Google, Apple and Citymapper. This can give useful insights into transport use patterns. The strong correlation between smartphone app data and actual trip data suggests policy makers could use smartphone app datasets in place of actual trip data, which could be useful in jurisdictions where the costs of gathering trip data are prohibitive. For example, it is estimated that by providing its own transport data to private companies, such as trip-planning app developers, Transport for London helped generate an extra USD 20 million (GBP 14 million) per annum in the wider supply chain and consumer spending gross value added while also helping to improve the efficiency of London's transport network through better mobility choices.

India also operates a mobile app which allows consumers to assess the energy efficiency, energy and monetary savings potential of different appliances as well as an option to provide product-related feedback.

In Canada, NRCan recently used a health and well-being smartphone app, Carrot Rewards, to gain insights into the audience and improve its EnerGuide home energy rating programme. The programme was initially launched by Canada’s government to advise citizens on energy-efficient products, vehicles, and home appliances and energy systems. Insights gained provided guidance on how to make information more accessible.

To expand the deployment of more energy-efficient electric vehicles (EVs), the Canadian government unveiled the Torque network app. The app allows current and prospective EV owners to connect with one another and with experts to share information, learn about key EV concepts and help create support for the uptake of EVs by building a community of like-minded people who can help to promote the benefits of the technology.

The Australian government also offers several online platforms that encourage citizens towards more energy-efficient choices. For example, the Energy Rating app calculates and compares home appliances’ energy performance, assisting consumers before they buy. In addition, the Light Bulb Saver app provides guidance on smart and cost-efficient house lighting options.

To boost behavioural change, the United Kingdom launched Greenwich Energy Hero, a digital service that installs smart meters to track energy usage in households. The app notifies users when electricity production is low in their area and rewards them for reducing their consumption in such hours. The 12-months trial ends with a comprehensive summary of the yearly energy and CO₂ emissions savings.

Natural language processing

With the help of natural language processing (NLP), machine learning and artificial intelligence, a large amount of text-based information, including from the internet, social media, transcribed interviews or focus groups, can be processed to extract critical themes, clusters and key messages. For example, the Government Technology Agency of Singapore and the Smart Nation and Digital Government Office have developed Ask Jamie, a virtual assistant with an embedded NLP engine. The artificial intelligence continuously assists citizens in their queries to Singapore’s government, connecting policy makers with end users on a wide range of issues.

The US Department of Energy has explored using NLP tools to scan through texts and numerical data on energy investments and company information to track innovation and clean energy progress. Outcomes include the identification of clean energy zones which can assist with policy targeting and design. The Environmental Defence Fund has also used NLP systems to analyse oil and gas permit applications that companies submit to regulators.

In Mumbai, NLP tools were used to help inform policies around the Slum Rehabilitation Housing programme to shed light on a range of energy poverty, social and gender-related challenges that arose with the programme. For example, it showed that new dwellings and lifestyles were more energy- and money-consuming, putting financial stress on occupants. Such digital tools can be used in a range of contexts, providing an important source of information for improved monitoring around the impacts of energy efficiency policies also helping give a voice to vulnerable groups.

Web search analytics

For appliances, regulators have typically conducted in-store audits or purchased sales data to assess if retailers are complying with minimum energy performance standards (MEPS). Web scraping provides an alternative method for regulators to assess whether models are being sold that do not meet MEPS. Using automated tools, regulators can quickly scan online shopping websites to assess which models are being offered for sale in their country. This was the approach taken in the NordCrawl project funded by the Nordic Council of Ministers and carried out by the Nordic countries (Sweden, Denmark, Finland, Iceland) during 2015‑2017.

The IEA has since co‑operated with NordCrawl in trials undertaken in Indonesia and South Africa, scraping through the countries’ online appliances market retailers. Results showed opportunities to improve information quality for consumers in terms of categorisation, and lack of comprehensive energy efficiency information. Similarly, the Premise Platform, in collaboration with the IEA, collected product data about air conditioners and refrigerators in retail stores in Indonesia, Thailand and Viet Nam. The findings were then leveraged to mould energy efficiency policy recommendations for the designated regions.

Analysis of web search data provided by services such as Google Trends can also inform policy makers’ view of markets for energy-using technologies at a lower cost than traditional market surveys or purchasing sales data. Such data also allow policy makers to compare trends in other countries. In emerging economies, where data on consumer trends are harder to obtain and market-monitoring resources potentially limited, web search data could prove a valuable resource for predicting market trends for energy use. In Chile, researchers found that models incorporating Google search data were able to accurately reflect consumer uptake of automobiles.

One way of testing whether a rebate for energy-efficient appliances is having the desired effect—encouraging people to purchase more energy-efficient goods—would be to analyse web search data or mentions of the programme or specific technologies in social media posts. This could be particularly useful where incentives are provided during a limited time period, such as tax-free holidays for energy-efficient goods, such as those offered in Maryland and Texas. While sales data would be ideal in these circumstances, analysis of terms such as “ENERGY STAR weekend” or mentions of specific appliances for which the incentive applied in the lead-up to and during the period of the incentive would provide at least a basic indicator as to the reach of the incentive.

The IEA has identified the Readiness for Digital Energy Efficiency (RDEE) framework, a broad set of policy issues that governments need to consider when they are seeking to increase the use of digital technologies for energy efficiency. This framework provides important guidance on what physical, social and regulatory infrastructure needs to be in place to enable policy makers to take most advantage of digitalisation.

The following issues should be considered by governments looking to expand their use of digital tools:

• Experience from Chile around smart meter roll-out found that building a positive digital mindset among citizens was key to the programme’s performance. Participatory approaches that engage all stakeholders involved and allow time to adapt can be helpful fostering consumer acceptance.
• For example, programmes encouraging the uptake of digital technologies and tools need to address cybersecurity and privacy concerns in order to earn users’ trust and confidence. Clear communication is needed about what data are collected, who has access to them, how they will be used, and how they will be stored and protected.
• Data collections should be paired with the establishment of management and protection systems for data ownership, privacy and sharing to build trust. Utilities may buy servers and run them themselves or elect to partner with companies whose core business is managing and operating large data systems with large international cybersecurity teams. Understanding cyber risk needs careful communication management.
• The effective use of digital technologies for energy efficiency requires timely data that are cleaned and anonymised before being made accessible to policy makers.
• Markets require access to high-resolution electricity demand data and meter-based quantification of changes in consumption as a result of energy efficiency investments. Adopting or expanding performance payments and programmes can help foster markets for energy efficiency and demand flexibility outcomes. Allowing competitive procurement can also bring programme costs down.
• Open-source digital tools for energy efficiency can be a powerful driver for technology and business model innovation, offering a valuable tool for fast and cost-effective uptake of new business models and regulatory frameworks. While multiple approaches to measuring the benefits of energy efficiency may offer some advantages as the best approach is determined, moving towards common metrics will be a powerful force for establishing strong markets in energy savings and other benefits. Agreeing protocols to promote consistency and harmonisation for distributed energy resources can help build market confidence and promote investment.
• The digital transformation brings both advantages and disadvantages to existing workforces and businesses. On one hand, it can open up much larger markets for energy efficiency, but on the other hand non‑digital methods and business practices may become obsolete. Signalling policies in advance and allowing time for businesses and individuals to adapt through the acquisition of digital skills can help lower the social costs of the transition. New digital programmes can be designed with this in mind.
• Care needs to be taken that benefits of using digital tools for energy efficiency also support vulnerable communities. Equity considerations such as alleviating energy poverty and promoting gender equality are essential goals for policy makers. Digital technologies offer innovative tools to better target and support such equity goals. However, this is not likely to happen automatically and there is a risk that digitalisation may widen inequality, unless the right incentives are put in place through policies.