The Acceptance of energy monitoring technologies: the case of local prosumers Mary Barreto Lucas Pereira Filipe Quintal ITI/LARSyS/prsma.com ITI/LARSyS/prsma.com ITI/LARSyS/prsma.com Funchal, Portugal Funchal, Portugal Funchal, Portugal Email:mary.barreto@prsma.com Email: lucas.pereira@prsma.com Email: filipe.quintal@prsma.com Abstract—With transformations happening in the electricity viously available to electricity suppliers become accessible to sector, we need to ensure consumers have access to updated and these consumers. However, there is some work to be done correct information to accompany such changes. Consumers need until that point, since they seem to have limited access to to understand technologies available to them but also, learn how to use them to optimize their personal investment in such types information about ways to optimize their current installations, of equipment. In this paper, we explore how a group of local pro- whether through upgrades or storage solutions that would be sumers has adopted energy monitoring technologies, their day- most adequate for them [4]. to-day strategies, and expectations when handling such systems. Energy monitoring systems address part of that issue, as We studied 11 prosumers and the technologies they have used these provide users access to the information being collected for three years, evaluated their satisfaction with the feedback provided and analyzed how a more modern visualization of their and consequently, enable them to act on their energy consump- energy practices was introduced and adopted into their daily tion. Energy monitoring technologies or eco-feedback are de- lives. We conducted interviews and questionnaires to evaluate fined as the use of technologies to monitor energy consumption their engagement with these tools. This initial work suggests this and production of a specific household [5]. These technologies particular group of users have already a high level of knowledge allow the collection of energy consumption and production about their systems, and as a result have integrated these into their routines. However, more support would be needed from by using sensors that send information from the household other local actors to help them reach more benefits and as such, to a central system. The information is gathered and then more satisfaction as consumers. We conclude by reflecting on presented using various forms such as email, or applications barriers that need to be addressed to increase user satisfaction that allow consulting and analyzing the information. In fact, with these systems. research studies [5], [6], [7], [8] highlight the importance of the mobility aspect. I. I NTRODUCTION In this paper, we report on a study of technology acceptance The increasing global population has lead to an equal of energy monitoring technologies with residential prosumers increase in energy use. Due to limited energy resources such as using small-scale solar PV to generate electricity. This study fossil fuels and, climate effects, energy efficiency has become describes participants’ motivations to become prosumers, the a challenging and pressing problem across the planet [1]. Since acceptance and usage of the PV system and PrsmaEnergy a large portion of energy use has been in the residential sector, the application developed within the project, and differences measures implemented in countries have focused on the energy between these two. The ultimate goal is to capture and efficiency of the existing buildings. Policy measures have analyze the empirical data with these users of how they have encouraged investments in innovative solutions such as smart used a PV system, observe how they adopted it through a grids, smart homes, self-generation, and storage technologies, more modern application installed to optimize it and identify which are available but still not widespread [2]. Photovoltaic potential problems that may exist while using it. Finally, systems (PV systems) for instance, are technologies that provide recommendations that can help service providers and operate in the area of clean energy, not focused on inefficiency the design of effective user-driven solutions. but replacing the energy source for renewable energy. It is a solution that helps consumers generate their own energy, and II. R ELATED W ORK it is easy to integrate within existing buildings. An estimate Due to the diversity of existing equipment that can be used of 17GW solar PV has been installed in the European Union, to achieve energy generation, there does not seem to be a and it is predicted to increase up to 32GW by 2030 [2]. precise term for residential prosumers. Generally, these are Consumers who use such systems are known as prosumers, known as energy consumers who produce their own energy energy consumers who also produce and generate their own but there is not a clear term to classify these users across energy. With the increased adoption rate and the technology countries due to differing regulatory frameworks and policies innovation of these systems, these consumers transition very [9]. Even within the EU there are no standard policies followed quickly from passive to active participants in the changing by the member countries that help create a set of incentives energy market [3]. In other words, decisions that were pre- to increase the adoption rates of PV systems. Although, it is expected for the installation of such systems to double changes emerge within the households. These changes are part by 2030, there are barriers that endanger such adoption [2]. of what researchers have designated as “demand shifting” [14], For instance, many consumers decide to install such systems where consumers change their behaviors to match and use because they see these as an extra source of income, whether most of the energy produced at the time it is being produced through savings in electricity or earnings, they realize that [13]. Constanza et al. [15] used an agent-based system to study producing their own energy could be very valuable, but when how users would shift laundry routines based on the electricity they search for the information, they realize local regulations tariff, and its results suggested increasing user interaction prevent them from doing so. In spite of some countries around automated systems to take more advantage of them. have already developed remuneration schemes, few enable Hansen et al [12] conducted a study with 20 households in prosumers to sell their electricity [2]. Consequently, reducing Denmark where they observed how participants changed their the uptake of users investing in PV systems. practices after having smart grid technologies installed. Results indicated participants became prosumers as they changed their A. Prosumer Motivations knowledge and behaviors to make the most use of the sun Recent work suggested a variety of motivations that explain while requesting more support and dialogue with the local why regular consumers decide to produce energy in their own energy provider. Similarly, Pierce et al [16] and Smale et al homes. At an initial phase, the financial factors determine [17] conducted interviews with users participating in smart the purchase of PV systems, for instance the upfront cost of grid trials to find cleaning practices were the ones most installation, borrowing costs, the scale of the financial benefit suitable for DSM, while other activities (leisure, cooking and (in terms of reduced electricity bills and available policy eating) were more limited to being conducted in other times support), and, the expected rate of return, earnings and savings than the ones usually taken to perform these. obtained (and payback period) of their investment [2], [10], Nevertheless, having access to information about energy [11]. generation is not always possible or easy to achieve. Besides In some instances, consumers relied on technical reasons the presence of other factors, such as social influences within to purchase a PV system, for example, ownership of an the household, there are technical limitations that prevent electric vehicle, smart meters or the ability to include at a prosumers from pursuing this demand shifting. later stage battery storage and demand response technologies. This paper describes a case study of a group of prosumers Other motivations included the aesthetics of rooftop solar PV located in a remote island, using energy monitoring technolo- panels, the security of supply, and finally, maintenance costs gies that keep track of their energy generation systems and and efforts over time. the technology acceptance observed and measured through the The environmental impact and commitment remain a moti- period of 4 to 6 weeks. Barriers are listed and analyzed in vation once the financial factors have been considered, and it terms of recommendations to improve such tools for future is based on the desire to protect future generations, and help service providers. reduce the environmental impact of fossil fuel usage [11]. The prevalence of some motivations over others is not III. S TUDY D ESIGN similar across European countries, due to the varied policies This study is integrated in the SMILE project that aims to being followed [9], nevertheless, financial factors and access to demonstrate nine different smart grid technologies in three capital seem to be the most prevalent. Becoming a prosumer, islands across Europe in order to foster its market introduction. particularly installing a PV panel, often starts with financial This study was conducted within one of the pilots (Madeira factors, but in a large number of cases it is a decision-based island) in the project with the main goal to smarten up the dis- in green and environmental motivations, and it becomes an tribution grid, through the optimization of self-consumption of extension of their lifestyle [12], [13]. renewable energy in the installations by introducing elements like Battery energy Storage system (BESS) and specialized B. Shifting energy activities battery management software. These prosumers are not cur- Demand-side management (DSM) has been one of the rently allowed to inject the excess energy into the grid due to strategies within energy efficiency initiatives, used to reduce local regulations, however, for the purpose of battery analysis or optimize the end user’s energy consumption in order to and testing, these will be allowed to do it. reduce the cost and the environmental impact. One of its One of the aspects of the project refers to the evaluation of mechanisms is the Demand response (DR) designed with the the energy monitoring technologies being used by participants. objective of adjusting production demand. It does so, through We started by recruiting prosumers in the island that owned PV the creation of dynamic tariffs that encourage consumers to solar systems and collected baseline information about their conduct their energy consuming activities spread over time usage of those systems. Around 4 weeks later, we installed a or throughout the day, consequently avoiding consumption monitoring system that collected baseline data for 6 months. peaks. From this demand-side perspective, prosumers are given Afterward, prosumers were given access to an application the choice of services they want to use to better match their that provided feedback on their energy consumption and needs, and which ones they will want to offer [3]. Once users production designated as PrsmaEnergy. This work describes have installed energy generation systems a set of practices or the technology acceptance evaluation of both the PV system (prosumers had already installed) and the project application PrsmaEnergy (which was integrated into the project). The acceptance was measured using: semi-structured interviews and a questionnaire adapted from the Technology Acceptance Model (TAM) [18], [19], [20]. The interviews collected information related to electric energy consumption and production, routines, expenses and awareness about energy-related habits or strategies. All inter- views, before receiving PrsmaEnergy and after using it for 4 to 6 weeks, lasted around 20 to 30 minutes. All interviews were recorded and transcribed, and its data were analyzed using grounded theory [21], [22] by grouping quotes into themes, Fig. 1. A screenshot of the “analytics” page comparing data about production and coding these according to categories, which are further and consumption from two different days. The user can select which day he would to see using the top right search fields. explained in the coming section. The Technology Acceptance scale was used before provid- ing access to PrsmaEnergy, and around 4 to 6 weeks after users addition, participants were asked about: frequency of use, most were given the access credentials to use it. The scale included relevant information and tools used to access it, least relevant 32 items across 6 dimensions: perceived usefulness (PU) (9 information and suggestions. After the interviews participants items), perceived ease-of-use (PEU) (7 items), intention to use were asked to fill in the acceptance questionnaire based on the (2 items), user satisfaction (7 items), ease of learning (4 items), TAM. and attribute of usability (3 items) (see table I for more details). The items were slightly adjusted to evaluate the acceptance of A. Participants energy monitoring technologies. It is believed the higher these The study was conducted with 11 households (see Table aspects are rated the higher is the user’s acceptance. The scale II) that were recruited through the project using information was translated to the native language of the user sample and sessions and ads in social media. During the recruitment users were asked to rate their degree of (dis) agreement on 7 phase, we encountered a combination of users that reported points Likert scale. The internal consistency of the scale was they wish to optimize their PV panels as a reason to get assessed by calculating the Cronbach alpha value of the overall involved in the study. All participants had a solar PV system scale, which was 0.98 for evaluating the current system, and installed in their homes purchased with their own funds with 0.76 for the PrsmaEnergy evaluation. no financial programs or subsidies using local companies as This study was designed to evaluate the technology accep- installers or recurring to online sales and installing these tance in two moments: themselves. Participants were asked to complete a baseline 1) Evaluation of the current PV system: Understand how survey that provided both demographic information and con- participants were using the PV systems they had already sumption practices. In terms of demographic information, the installed in the homes before the project even started. The household size ranges from 2 to 5 people, with an average research team conducted semi-structured interviews to collect of 3.54 people per household. Age ranges of participants and information on the following aspects: family members vary between 2 years old and 84 (average 1) The reasons and motivations for purchasing such a age is 37.3 years old). We measured participants environmental system and becoming a prosumer; concern by asking them to fill in the New Environmental Scale 2) Expenses before and after, as well as current ones; (NEP) [23], to which all responded in agreement with the 3) Strategies used to optimize their production; new environmental paradigm. All participants were supportive 4) Routines and changes; of the impact of human activity in the environment and its 5) Use of an application that provides information about fragility, higher scores of this new paradigm are associated their energy production, in case they had one. It included with pro-environmental behaviors [23]. a Think-Aloud while using the PV system. After the in- terview participants were asked to fill in a questionnaire B. PrsmaEnergy set up based on the TAM [18], [19], [20]. Each participant household was already equipped with a 2) Evaluation of the PrsmaEnergy satisfaction and usage: PV panel to which a smart meter was installed to collect Understand how and in which ways PrsmaEnergy changed baseline data in terms of 1) energy production, 2) energy their behaviors or use of the PV panels. Participants were consumption, 3) energy being sent to the grid. This data given access to the digital platform and were interviewed was then stored in the projects Energy Management System 4 to 6 weeks after having used it. The interviews focused (EMS) and displayed through PrsmaEnergy designed with the on following up aspects from the interviews conducted in purpose of tracking the individual installations and providing the first moment, namely, changes in expenses, changes to information about their systems throughout the project. In their routines using devices or the system, and strategies terms of the feedback, it was decided to represent electricity they used to handle or interpret the information given. In production and consumption in terms of 1) energy and power, TABLE I C ONSTRUCT DEFINITION USED IN THE ACCEPTANCE QUESTIONNAIRE . Construct Definition Perceived usefulness The extent to which a person believes that using a technology will enhance her/his productivity Perceived ease of use The extent to which a person believes that using a technology will be free of effort Intention (to use) The intention to use the system as often as possible and as long as it is available User satisfaction The extend to which the user feels confident in using the system Ease of learning The extend to which the user finds it easy to learn and operate Attribute of usability The extend to which the user finds the system to be well integrated and easy to interact with TABLE II PARTICIPANTS DEMOGRAPHIC INFORMATION User ID Household size Ages Occupation (main caregiver) Time as a prosumer (years) A 4 44,42,14,11 Nature Watcher 18 months B 4 84,48,47,22 Teacher 1 year C 4 56,52,23,21 Retired 2 years D 3 42,41,6 Nurse 2 years E 2 74, 32 Unemployed 10 months F 4 62,32,02,44 Business 3 years G 3 51,45,20 Construction worker 3 years H 2 64, 64 Electrician 2 months I 4 55,50,20,15 Teacher 3 years J 5 42,41,13,10,7 Businessman 4 months K 4 55,55,28,19 Government worker 3 years 2) cost and 3) environmental impact. The metrics selected - kWh/kW, money, and CO2 - are indeed commonly used in combination for the specific purpose of engaging different target groups [5], [24]. The features included close to real-time feedback about consumption and production (updated every minute), historical feedback with three levels of temporal granularity (days, weeks, months), and comparison between dates the users could select to have more information. Real-time feedback (i.e. cur- rent consumption and production), together with an overview of recent energy use (last hour and current day/week/month), is displayed in the “dashboard” (which serves as a landing page), while the “analytics” feature provides users with the opportunity of reviewing and comparing data among different periods of time (Figure 1). We selected three main visualization types (Figure 2): lines chart (to show patterns over time - e.g. the period of the day when production usually exceeds consumption), pie and bar charts (to represent cumulative amounts of data - i.e. daily, weekly, and monthly energy use). Households usually value the opportunity to access feedback and check their energy status via mobile devices. For this reason, we opted for a web- based application designed to be responsive, so that it can be accessed also via mobile devices (smartphone and tablets). The work presented here aimed to evaluate the acceptance of energy monitoring technologies by a local group of prosumers, focusing on the following questions: 1) What motivations led participants to become prosumers? Fig. 2. The three main visualization types provided by PrsmaEnergy, the first 2) How do they report themselves in terms of electricity one displays consumption, production and excess energy. The second presents this data throughout a week. The bottom visualization provides details about usage and other resources? energy produced versus energy used from the grid. 3) What was the acceptance for the current PV system they had installed in their homes before we started our research project? 4) Were there differences in terms of previous system and B. User motivations to become prosumers PrsmaEnergy? All prosumers used their own means to install the solar 5) How did participants use PrsmaEnergy? PV system, meaning they used local installers, searched for The following section summarizes the findings for each local information, contacted the local electricity company of these questions, with a combination of qualitative and or used online sources. They mentioned this process to be quantitative results, gathered through the questionnaire and cumbersome, as there were no clear paths on how to proceed the interviews conducted with the participants. The quali- whether in terms of subsidies they could apply to or even tative analysis displays more relevant and common themes regulation they would need to follow. that emerged from the participants’ interviews concerning In terms of the reasons and motivations for purchasing technology acceptance. The quantitative results refer to the PV panel system, users reported this decision was aligned Acceptance scale used to evaluate both systems. to their willingness to keep up with the latest technologies or the opportunity to have an innovative system at their IV. R ESULTS home. In other words, they are the early adopters of such This section presents the key findings of the acceptance technologies and perceived their purchase as an opportunity study in three areas: a) users energy practices, b) user mo- to try something new in the area of renewable energy. tivations and c) technology acceptance, of the current system Regarding the environmental value users, viewed it as a and of PrsmaEnergy. For the first two, the main themes are way to protect the environment but also, part of a greater described in a succinct way, and for PrsmaEnergy, Table III goal, as they saw it as a part of a lifestyle they want to presents the main themes with illustrative quotes. build around environmental protection and preservation. Not A. Users energy practices only in the field of energy but also, food production, purchase of organic products or even sustainable home construction. When asked about their electricity costs before (the costs Whenever the environmental concerns were mentioned, users ranged from 75 to 130 euros) and after the installation of had already integrated into their lives these aspects beyond the the PV panel (the costs lowered from 45 up to 100 euros) area of energy, and these were measures they had started to all participants referred to a cost reduction between 20 and progressively adopt into their day to day living, even before 30 percent after installing the PV panels. Participants reported the self-consumption systems came into the market. the costs based on expense tracking habits they had before and Other users referred to the availability of solar energy on the after installing the PV panel. Four out of the 11 users reported island, they saw it as a valuable and ready to operate resource slight increases to their energy bill, which they interpreted as that was not being used to its fullest potential. For them, tax increases. Or even due to weather changes, for instance, the it seemed it could be even more capitalized and an obvious fluctuating temperatures during the summer required the use of choice to follow by our governments, by our local companies air conditioning or fans in the homes. One user shared the PV and supported by all kinds of programs. It was not clear to system did not allow so far savings, due to specific appliances them why this is not supported, as it should be. As well as in the home that did not allow for it just yet, namely, the water why are not there more consumers investing in PV systems pump used for agriculture. In spite of these increases, all users for their home. mentioned the decision to purchase such a system was a great A few users mentioned their motivation to have PV panels one; otherwise, their expenses would reach higher values. was associated with the need to having greater information To understand their routines and the use of electrical appli- about the electrical appliances they have at their homes, which ances, users were asked to list their daily use to which they consequently helps preventing future problems. Either because reported the following: they ran into device malfunctioning in the past, or had an • A set of devices is always used every day (coffee ma- energy surcharge that affected their electrical appliances, or a chine, microwave, and kettle); remote way to monitor their equipment when away from home • A set of devices is always used during the day when they for longer periods of time. This allowed them to keep a record have the highest energy production (washing machine, of information they can use as justification in case they have stovetop, oven, vacuum cleaner); a situation of malfunctioning, and they need to report it to the • When there is not enough production they postpone the local electricity company. usage of the device for the next day - for instance, doing The cost reduction was a least motivation mentioned, with laundry (washing machine) or cleaning (vacuum cleaner) time users realized achieving full autonomy from the local grid when it is sunnier; is still a long distant goal, however, they are pleased for being • In some cases, they anticipate the realization of some able to use the energy produced by their self-consumption activities as early as possible to avoid using energy from units during the day. the grid at night - for example, they cook dinner earlier than usual (they know the stovetop and oven are highly C. Technology Acceptance consuming); With regards to the acceptance evaluation, users scores were • They charge their phones during the day, which they tried higher for PrsmaEnergy to the exception of one user. To inves- to implement whenever possible. tigate further, we proceeded to test the differences between the two acceptance scores. Due to the data not following a normal • the lack of being closer to real-time activities. In other distribution for some variables, a Wilcoxon Signed-Ranks Test words, the system should be placed closer to the spaces was run and the output indicated the PrsmaEnergy acceptance where devices are mostly used such as the kitchen or scores were statistically significantly higher than the current easily available to other family members. In fact, the system acceptance scores Z = 26.5, p < .034. frequent user of the application in all homes is male, the Users scores were higher for all the sub-dimensions for person who decides to purchase the system, the person PrsmaEnergy (see Figure 3). The highest scores were on who is at most ease with technology and the one most Perceived usefulness, which means users felt the application interested in such equipment. made them believe it would enhance their productivity, or more 2) PrsmaEnergy: Regarding the application provided by specifically help them collect the most relevant information in the project (see Table III), users reported using it more a fast, easy and simple way about their energy production. frequently in the first three weeks either through their mobile The second highest scored was for User satisfaction, meaning phone or their personal computer. When they wanted to take prosumers felt confident using the application, felt it was a quick glance at the data they would use their mobile phone, working as expected and they would recommend it to a friend. while if they wanted a more detailed analysis they would use Intention to use was the lowest for both the PV system and the computer to enlarge the graphs. Once they learned how to PrsmaEnergy. interpret the production information based on their household In the following two subsections, we explore even further routines and/or weather conditions they used it less frequently. the results regarding the PV system and the project application, In terms of reasons to interact with PrsmaEnergy, they through a summarized description of users perceptions of both would use it to keep track of their energy production informa- systems. tion and keep track of their savings. All users mentioned this 1) PV system: Users were asked about strategies used to reason several times. Some users were more specific in terms optimize their production once they installed the system, and of which moments led them to use the application, such as: these included the use of devices such as washing machines, dishwasher or dryers during the day. For that, they would either • to check the influence of a new appliance; program the machines to be used in the highest sun peak hours, • to confirm production values according to the weather or stop using these after a certain time in the day, for example, conditions on that specific week or day (especially for avoid using these after 4 PM in the winter and 6 PM in the sunny days); summer. Another strategy was to purchase timers to control • to assess if certain activities could be conducted using device usage, especially at night to regulate their energy usage self-consumption; from the grid. The overall goal was to avoid using energy from • to identify energy peaks that could be avoided to optimize the grid, as a result, they specified to use as few devices at even more their system; night as possible, replaced devices that were demonstrating • and to assess the feasibility of adding other renewable high consumption for more efficient ones or even replaced energy systems to his home. some lighting to more efficient lamps. Users mentioned the most relevant information to be the In terms of improvements, they would like to conduct in money being saved while using most of their energy pro- their installations these would be the purchase of storage duction, the carbon dioxide emissions and the production solutions to use the energy they know is not being used and information they used as a hint to start certain electrical currently wasted, however, the high costs and the lack of appliances in order to avoid using energy from the grid. financial support programs prevent them from making such In terms of routines, users felt PrsmaEnergy was a good tool investments. As well as the lack of trustworthy information of to plan laundry and to analyze consumption peaks in order reliable systems they can use to optimize even more the use to know where and when they could further optimize their they make out of their systems. efforts, whether by moving one activity to another time of the Regarding the use of an application with information about day or stop using one appliance in case there were too many their production, only 7 out of the 11 users mentioned to have being used. One user mentioned the information was helpful to it available everyday or rather frequently. The other 4 users make the family aware, however, they felt they were not taking did not have access to their energy production information. it to the next level and making these into actual strategies to The users who did use it did it to plan their activities in the change their practices at home. Users continued to do the most homes, and their device usage based on that information. In consuming activities during the times when there was more addition, they mentioned the information was useful to: energy being generated by the PV system. • detect consumption peaks; In terms of suggestions one user who has batteries installed • remotely control their homes when away; in his home, would like to have had its information added to • and to plan their household activities. PrsmaEnergy, similarly to the PV system application that was In terms of weaknesses, users mentioned: included with it. • the complexity to sometimes customize the visualizations, Users found the application user-friendly and accessible in to the extent they had not figured these applications quite some cases even more than the PV current system. In terms well; of family members using the application, the majority of users Fig. 3. Technology acceptance for each system in terms of sub-dimensions TABLE III P RSMA E NERGY THEMES Themes User quotes examples Production Information Now I really want to buy the (wind) turbine to see through the application how much energy can I produce when there are windy days and perhaps take advantage of night periods to collect it (User C) Triggers to use it I look at the production and the consumption to make sure everything is working (User D) I was not aware of the best time to start the washing machine in the morning and now I avoid starting it too early such as 9 am because there is not much sun yet, it is better to start it after 10 or 11 am, it’s the time to start using more appliances (User E) I use it mostly when it is sunny, and I’ll check several times on those days, when it’s rainy weather I don’t even bother to look at it because I know it is not producing as much energy (User F) The most relevant for me is seeing the bill being reduced, the more savings the better (User K) Most relevant information I find it very interesting that you have added how much carbon dioxide am I responsible for, I find it very interesting and it was a very great choice from your side to include it in the feedback (User I) Ease of use Your application is much better, more reliable more complete than the one I used to use that came with the system (User C) I think it is very good, it gives the necessary information for regular people who are not experts, if it had much more information I might not know how to interpret it or what it meant. It is simple and user friendly (User D) The information I need is here, it is easy to see and very intuitive for me (User J) mentioned being just themselves checking the application and was installed. In fact, the application enabled them to infer using the information in their daily lives. In spite of their even more information, such as adding or upgrading their efforts to captivate their attention to look at the information, current PV system with other energy generation technologies. or told they should avoid using an appliance due to low energy This follows the recommendations by [2], where consumer production, other family members would not use it. acceptance can be increased if governments and local munic- ipalities support the growth of complementary technologies V. D ISCUSSION (Electrical Vehicles and storage technologies). Participants in this study being prosumers revealed high Users continued to use PrsmaEnergy to plan activities and scores in terms of environmental concern and had their energy detect consumption peaks. The activities more prone to be use transformed by the decision of installing a PV system. shifted were also cleaning and laundry, such as [17], but also, Similarly to [12], they also changed their consumption activi- with some prosumers the activity of cooking was considered ties to take advantage of the hours with most energy produc- due to their flexible working schedules. However, in both tion, with the current PV system and even after PrsmaEnergy systems, the most frequent users were the person responsible for purchasing the PV panel, for agreeing to be part of the protection mindset. study, and the one that would already use the information. In terms of features used, users referred mostly to the most The fact that other family members were not mentioned, in recent information, and hardly to historical information, which the majority of the cases, it means PrsmaEnergy could be can be explained by the fact that users would use the presence further improved to engage other users within the homes, of sun as a trigger to use the application. The fact that this since practices tend to be influenced by the social dynamics study occurred in the Winter, they felt less motivated to use the of families, which has been seen in the area of eco-feedback application and conduct comparisons between dates, meaning [25] but also, energy production [13]. they would only interact with it while there were sunnier In what concerns motivations to become prosumers, our days and less interested in exploring different production days. users contrary to what authors have found so far [2], [11], Users enjoyed the application because it expanded on the hardly mentioned financial concerns or earnings. At least, this lifestyle they are trying to pursue, by having information on did not emerge until PrsmaEnergy displayed such information, the emissions they felt it came to strengthen their motivations but only for two users out of the 11. The remaining users to use as much renewable energy as possible. Overall, partic- focused mostly in using the latest technology, helping the ipants revealed high scores in terms of technology acceptance environment and that it could be even further used, but local for both the PV system they installed and PrsmaEnergy that authorities or organizations do not offer proper incentives for came to increase their satisfaction with the user experience more people to join the green energy movement. Although and information provided. users did not verbalize the financial concerns being a major motivation, we believe that beyond the green values and VI. R ECOMMENDATIONS environment protection, prosumers revealed a need to have greater control over their electricity in their own home. Since Prosumers are a group of people with a considerable level they used both PV system and PrsmaEnergy information to of knowledge and are willing to invest their own resources in keep track of electrical appliance problems, but also problems order to optimize the equipment they have already installed resulting from the grid supply, and to further learn about their in their homes. The number of prosumers is likely to increase consumption and production practices. as the technologies become more accessible. Prosumers have Taking in consideration the fact that prosumers in this study motivations that start by being financial, but just partially as invested their own resources to pursue their motivations of the regulation in each country will support it or not. But this having an energy generation system, it seems that incentives does not stop them from conducting an upgrade to the system for this group of people and others in similar cases should go or even considering other energy efficiency solutions. beyond economical grounds. One way could be through the It is becoming more common to find regular consumers creation of subsidies to help users upgrade their installations, interested in green values and environmental protection, and but also, include the access to information campaigns, or the not just because of their local municipality awards it. On the creation of community-based activities that allow users to contrary, this study showcased a group of prosumers that are share information and experiences since the installations had pursuing their goal to become more innovative and become specific technical needs that could be overcome through shared more self-sufficient in spite of having no subsidies or local knowledge. programs to help them improve the current installations. This In terms of technology acceptance, users felt PrsmaEnergy project came to help them, what they hope to assess the feasi- was easier to interpret, more complete, hence the highest bility of potential storage solutions or even just the installation scores for the application. Consequently, users were more in itself. What we observed was that prosumers are more than satisfied with it as reinforced by their testimonies in the willing to invest in energy generation technologies if it means interviews. The high scores can also be explained by the having greater control of their costs, their information, their fact that all prosumers decision and investment for the PV consumption patterns and practices, as long as these work and system was their personal initiative using their own financial are user-friendly. resources, supported by the desire to have an innovative system How can ICT help these citizens reach a more sustainable and have the opportunity to experience it first hand. One user future? We believe ICT cannot work alone if local organiza- scored the PV system higher than PrsmaEnergy in terms of tions are not involved, and for that matter, ICT can help group acceptance, due to the fact that his own application included and gather the data and information in a more meaningful way in the equipment, provided the information about his storage in order for municipalities to create tangible measures con- equipment. This user has the most complete PV system in sumers can easily put into practice. The following measures the sample, and our application will only include this type of could be a start to this kind of work: information at a later stage in the project. As a result, his scores • create community digital platforms that consumers can are understandable, since the user expects an application to be access to have clear and neutral information about energy even more innovative compared to the one he already interacts generation systems and regulation; with on a daily basis. Nevertheless, this user was impressed by • disseminate local experiences with these technologies; the addition of the emissions information, which he believes • facilitate participatory discussions between governments, is important to bring more people into the environmental companies, and consumers about contextual conditions that could be implemented to optimize the local re- X. ACKNOWLEDGMENTS sources, and motivate integrated collaborations between This work was funded by the European Union Horizon 2020 all stakeholders for a greener mindset. research and innovation programme under grant agreement number 731249, by the Fundação para a Ciência e Tecnologia VII. L IMITATIONS under the grant agreement number UID/EEA/50009/2019 and by ARDITI - Agência Regional para o Desenvolvimento One limitation of this research is the fact this is a small da Investigação Tecnologia e Inovação through the support sample of participants that live in a specific context where they provided under the grant agreement number M1420-09-5369- are not allowed to inject the excess produced energy into the FSE-000001- Bolsa de Pós- Doutoramento. grid. This alone makes them prone to become motivated to use most of their energy production through a number of strategies. R EFERENCES Having an application that helps them to do so, explains their [1] L. Pérez-Lombard, J. Ortiz, and C. Pout, “A review on buildings energy high scores in terms of acceptance and satisfaction. Having consumption information,” Energy and buildings, vol. 40, no. 3, pp. conducted the study in a location where prosumers could make 394–398, 2008. [2] The European Commission, Study on Residential Prosumers in the direct earnings from their production, we are left to wonder if European Energy Union. The European Commission, 2017. the efforts would remain the same. Another limitation regards [3] Y. Parag, “Beyond energy efficiency: A ‘prosumer market’as an in- the fact that a large part of the savings before and after the tegrated platform for consumer engagement with the energy system,” eceee 2015 Summer Study on energy efficiency, pp. 15–23, 2015. system, were self/reported, even though participants did keep [4] S. Gährs, K. Mehler, M. Bost, and B. Hirschl, “Acceptance of ancil- an accurate track of their expenses, the study would have lary services and willingness to invest in pv-storage-systems,” Energy benefited if expenses analysis prior to our study had been Procedia, vol. 73, pp. 29–36, 2015. [5] J. Froehlich, L. Findlater, and J. Landay, “The design of eco-feedback analyzed, either through the utility company records or the technology,” in Proceedings of the SIGCHI conference on human factors participants. in computing systems. ACM, 2010, pp. 1999–2008. [6] G. Fitzpatrick and G. Smith, “Technology-enabled feedback on domestic energy consumption: Articulating a set of design concerns,” IEEE VIII. F UTURE WORK Pervasive Computing, no. 1, pp. 37–44, 2009. [7] L. Pereira, F. Quintal, M. Barreto, and N. J. Nunes, “Understanding This study is integrated into a larger scale project that aims the limitations of eco-feedback: a one-year long-term study,” in human- to maximize self-consumption installations into the current computer interaction and knowledge discovery in complex, unstructured, big data. Springer, 2013, pp. 237–255. grid. The next stage consists on splitting our sample of [8] J. Kjeldskov, M. B. Skov, J. Paay, and R. Pathmanathan, “Using mobile prosumers into two, one where prosumers will have a battery phones to support sustainability: a field study of residential electricity installed in their homes, and a second one, where prosumers consumption,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM, 2012, pp. 2347–2356. will have more detailed feedback based on their production [9] E. Heiskanen and K. Matschoss, “Understanding the uneven diffusion of characteristics. The goal will be to compare the optimiza- building-scale renewable energy systems: A review of household, local tion strategies used within the two samples. The technology and country level factors in diverse european countries,” Renewable and Sustainable Energy Reviews, vol. 75, pp. 580–591, 2017. acceptance will be evaluated again, with the introduction of [10] The European Commission, Consumer surbey 1 - Attitudes, opnion, batteries and its usage in PrsmaEnergy, with an added level of drives and barriers and satisfaction with regard to Renewable Energy feedback. This will be essential to understanding acceptance Systems. The European Commission, 2014. [11] P. Balcombe, D. Rigby, and A. Azapagic, “Motivations and barriers and satisfaction for this type of systems, and which factors may associated with adopting microgeneration energy technologies in the uk,” affect its user experience or satisfaction, and market entrance. Renewable and Sustainable Energy Reviews, vol. 22, pp. 655–666, 2013. [12] M. Hansen and B. Hauge, “Prosumers and smart grid technologies in denmark: developing user competences in smart grid households,” IX. C ONCLUSIONS Energy Efficiency, vol. 10, no. 5, pp. 1215–1234, 2017. [13] J. Bourgeois, J. Van Der Linden, G. Kortuem, B. A. Price, and C. Rim- In this paper, we have presented the technology acceptance mer, “Conversations with my washing machine: an in-the-wild study of demand shifting with self-generated energy,” in Proceedings of the of energy monitoring technologies, namely solar PV systems 2014 ACM International Joint Conference on Pervasive and Ubiquitous and an application providing detailed feedback on energy Computing. ACM, 2014, pp. 459–470. consumption and production designated as PrsmaEnergy. The [14] J. Pierce and E. Paulos, “Beyond energy monitors: interaction, energy, and emerging energy systems,” in Proceedings of the SIGCHI Confer- groups of prosumers that participated in this study are a group ence on Human Factors in Computing Systems. ACM, 2012, pp. 665– of people that are deeply motivated to change their practices to 674. optimize even more the technology they have installed, how- [15] E. Costanza, J. E. Fischer, J. A. Colley, T. Rodden, S. D. Ramchurn, and N. R. Jennings, “Doing the laundry with agents: a field trial of a ever further support to expand such systems would be needed. future smart energy system in the home,” in Proceedings of the SIGCHI The interviews with the participants unraveled their challenges Conference on Human Factors in Computing Systems. ACM, 2014, rely mostly on getting other family members to get involved pp. 813–822. [16] J. Pierce and E. Paulos, “The local energy indicator: designing for wind and obtaining governmental support, their expectations are and solar energy systems in the home,” in Proceedings of the Designing to improve even further their systems with other generation Interactive Systems Conference. ACM, 2012, pp. 631–634. technologies, but also storage. PrsmaEnergy expanded their [17] R. Smale, B. van Vliet, and G. Spaargaren, “When social practices meet smart grids: Flexibility, grid management, and domestic consumption in knowledge and likewise allowed them to identify more strate- the netherlands,” Energy Research & Social Science, vol. 34, pp. 132– gies to optimize even further their systems. 140, 2017. [18] M. P. Gagnon, E. Orruño, J. Asua, A. B. Abdeljelil, and J. Emparanza, “Using a modified technology acceptance model to evaluate healthcare professionals’ adoption of a new telemonitoring system,” Telemedicine and e-Health, vol. 18, no. 1, pp. 54–59, 2012. [19] V. Venkatesh and F. D. Davis, “A theoretical extension of the technology acceptance model: Four longitudinal field studies,” Management science, vol. 46, no. 2, pp. 186–204, 2000. [20] H. M. Abu-Dalbouh, “A questionnaire approach based on the technology acceptance model for mobile tracking on patient progress applications,” Journal of Computer Science, vol. 9, no. 6, pp. 763–770, 2013. [21] A. Strauss and J. Corbin, “Grounded theory methodology,” Handbook of qualitative research, vol. 17, pp. 273–85, 1994. [22] J. Corbin, A. Strauss et al., Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage, 2008. [23] R. E. Dunlap, K. D. Van Liere, A. G. Mertig, and R. E. Jones, “New trends in measuring environmental attitudes: measuring endorsement of the new ecological paradigm: a revised nep scale,” Journal of social issues, vol. 56, no. 3, pp. 425–442, 2000. [24] C. Fischer, “Feedback on household electricity consumption: a tool for saving energy?” Energy efficiency, vol. 1, no. 1, pp. 79–104, 2008. [25] M. L. Barreto, A. Szóstek, E. Karapanos, N. J. Nunes, L. Pereira, and F. Quintal, “Understanding families’ motivations for sustainable behaviors,” Computers in Human Behavior, vol. 40, pp. 6–15, 2014.