Skip to main content

NATural Language Energy for Promoting CONSUMER Sustainable Behaviour

Periodic Reporting for period 2 - NATCONSUMERS (NATural Language Energy for Promoting CONSUMER Sustainable Behaviour)

Reporting period: 2016-05-01 to 2017-06-30

Residential energy consumption represents the 28% of all EU consumption and if commercial buildings are also considered this percentage increases to 40% (36% of EU CO2 emissions). In this context, is clear that the reduction of consumption in the residential sector should play an important role in energy efficiency programs and policies as is stated in the Energy Efficiency Directive 2012/27/EU. Most energy efficiency measures implemented in Europe involved technological interventions. In contrast, everyday energy-consuming behaviors are largely habitual and therefore the potential of energy savings at home with actions focused in consumer behavior is really promising.

NATCONSUMERS is a project designed to improve feedback to consumers about energy usage within the European residential sector. The project have developed a methodology for communicating more effectively with consumers by using ‘Natural Language’ – communication which is friendly, emotionally intelligent, relevant and simple. This advice would be designed to raise awareness about how people use energy within their homes and to give them advice about how to use energy more sustainable. With the role out of smart metering across Europe, the project looked at incorporating smart meter data to provide advice to householders.
Tailoring of energy advice has been found, through multiple studies, to be much more effective than more general advice provision. Variations in consumers’ characteristics and behaviors lead to heterogeneous energy demands, influenced by both individual preferences and physical variables. Furthermore, framing advice in differing ways for different consumer can improve the likelihood of advice being acted upon. Therefore, NATCONSUMERS developed a feedback framework which would help energy utilities and third parties to send tailored messages to residential energy consumers, supporting them in increasing energy efficiency.
In the project, we made a lot of effort to gather all the existing knowledge in the literature and open a dialogue with PSC members and consumers about energy reduction possibilities.
In order to do this, we have arranged 3 workshops (Oslo, Madrid, Naples) and conducted several focus group in different EU countries (Spain, Italy, Norway, Hungary). During the project, over 100 academic articles were reviewed, and the most important factors were identified which influences energy consumption. As second step a survey questionnaire was developed and large-scale consumer surveys were conducted in four countries of the EU, in Denmark, Hungary, Italy and the UK. Survey results were used to elaborate the attitudinal segmentation needed for sending tailored messages to consumers. 

The underlying premise which NATCONSUMERS is built upon is that for energy efficiency advice to be truly effective, it must be tailored. In order to achieve this, within NATCONSUMERS we have created three segmentation models. The first utilises smart-meter data to categorise consumers based upon their electricity load profile – i.e. based on their patterns of energy usage over time. The second segmentation is based on socio-demographics. This has been used to investigate how much electricity households use, or the total ‘volume’ of consumption. Combined, these two segmentations allow us to paint a picture of a user’s overall energy usage, in terms of both patterns and quantities of use. 
The third segmentation has been constructed from a survey of consumers’ attitudes and values. From this, we can identify what interests each householder and can therefore re-frame the message in terms that they will take note of. This segmentation therefore identifies how we should communicate with each individual. 
These three segmentation models have been used to build the NATCONSUMERS Natural Language Generator. This generator inputs smart-meter data and consumer attitude data, and outputs Natural Language energy efficiency advice. Three consortium members each took on the task of helping to develop a corpus of energy feedback. Each member ran a set of corpus development workshops each with a slightly different approach.
Based on the results obtained throughout the project a framework structure was created based on the natural language point of view. The objective of this activity was to generate a step-by-step guide that illustrates the complete process for the generation of energy saving messages using natural language. 

When developing NATCONSUMERS advice, an appropriate mechanism for delivering the messages must also be developed or selected. These potential mechanisms are described as engagement concepts. Based on the information extracted from workshops and from the complete characterisation of energy consumers and their requirements, a creative session was held in September, 2016 in London.
The Design Jam generated around 30 ideas from which 11 consolidated concepts were extracted. Each concept was evaluated based on a certain number of criteria ranging from the long lasting effect of the solution, its potential impact in changing behaviour, its technical feasibility, its level of user-acceptance, etc. In order to assess and evaluate the 11 engagement concepts resulting from the Design Jam, we conducted ten focus groups across Italy, Hungary and Norway. 

NATCONSUMERS was envisaged from the start as a highly participatory project and, through our Project Stakeholder Community (PSC), engages with numerous actors in the energy efficiency value chain.  After the first year of the project, the PSC is made up of 50 professionals from 39 organisations from 17 European countries. During the second year of the project, the interactions between the PSC and the project partners took place as follows: Two new members were added to the PSC; holding a webinar, review the handbook, active participation in the final conference, working together on future exploitation. 
NATCONSUMERS incorporates several innovative elements that go beyond the state of the art from a technological and energy perspective.
- Natural language (NL) feedback approach
- Tailored and relevant energy saving advices
- Provision of direct energy efficiency action to the consumers
- User-centred approach
- Advanced, integral characterization and segmentation of EU energy consumers

To measure the impact we defined a set of Key Performance Indicators (KPIs) and built a simulation model to evaluate the performances of the tool for generation of feedback
The wide savings range for all categories shows that consumer's understanding and use of feedback is one of the most important factors in feedback programs. The modelling shows that these benefits are more substantial for certain profiles of users, generally the most energy-intensive users.
Aggregated individual savings arising from feedback also have an impact on the power system at national level. A lower national demand for electricity (both throughout the day and during system peak hours) leads both to a lower request to generate electricity (reduction in fuel consumption) and a lower need to import electricity. Finally, from the environmental point of view a reduction in fuel consumption for generation plants means a reduction in CO2 emissions. Logically, the more household participate in feedback programs the greater the impact, hence the importance for market actors to focus very much on customer acceptability of feedback offerings.