Building Occupants, Their Behavior and the Resulting Impact on Energy Use in Campus Buildings: A Literature Review with Focus on Smart Building Systems

Abstract

In the light of global climate change and the current energy crisis, it is crucial to target sustainable energy use in all sectors. Buildings still remain one of the most energy-demanding sectors. Campus buildings and higher educational buildings are important to target due to their high and increasing energy demand. This building segment also represents a research gap, as mostly office or domestic buildings have been studied previously. In the quest for thermal comfort, a key stakeholder in building energy demand is the building occupant. It is therefore crucial to promote energy-aware behaviors. The building systems are another key factor to consider. As conventional building systems are replaced with smart building systems, the entire scenario is redrawn for how building occupants interact with the building and its systems. This study argues that behavior is evolving with the smartness of building systems. By means of a semi-systematic literature review, this study presents key findings from peer-reviewed research that deal with building occupant behavior, building systems and energy use in campus buildings. The literature review was an iterative process based on six predefined research questions. Two key results are presented: a graph of reported energy-saving potentials and a conceptual framework to evaluate building occupants impact on building energy use. Furthermore, based on the identified research gaps in the selected literature, areas for future research are proposed.

1. Introduction

In the EU, buildings are the largest energy consumers and one of the main CO₂ emitters [1]. Considering climate targets such as the European Green Deal and RepowerEU [2] that call for climate neutrality by the year 2050, including EU energy independence, a substantial shift in how buildings are designed, operated and recycled is required [3]. The increase in global energy use and depletion of natural resources also call for energy-saving strategies [4].

In the IEA EBC Annex 53 project, six parameters that can affect building energy use were identified: climate, building envelope, building equipment, operation and maintenance, indoor environmental conditions and occupant behavior. These six parameters were further classified into technical and physical factors (the three first) and human-influenced factors (the three latter). They were developed in order to enhance knowledge about building energy use and building energy data and they could also serve as support to detect potential energy savings [5].

Understanding and promoting energy-aware behavior in buildings are important aspects of the transition towards climate-neutral buildings. Building occupants’ impact on building energy use is twofold; as stated by the World Business Council for Sustainable Development (WBCSD), “wasteful behavior can add one-third to a building’s designed energy performance, while conservation behavior can save a third” [6] (p. 62). Instead of seeing the users as a problem, it is important to see them as part of the solution. Hence, the implementation of any energy-saving measures cannot fully reach their potential if the building users are not integrated [7].

Research agrees that the behavior of building occupants affects the building’s total energy use [8]. The literature mentioned several examples of energy-related building occupant behaviors that could lead to an increase in building energy use. For example, keeping the lights on when leaving a room [8,9,10,11,12], neglecting to turn off the HVAC system upon leaving a building or interfering with thermostats [9,12] and windows as well as door opening behavior [11,12]. Window opening behavior is especially problematic due to the fact that there will be an increase in dynamic heat losses due to an excessive amount of air flow; hence, more energy is required to heat incoming air [13]. Studying occupant behavior and its impact on building energy use is a complex matter that involves a multidisciplinary field of sociology, psychology, economics, engineering and design parameters [14]. Climate and outdoor temperature are two parameters that impact the behavior of building occupants, often resulting in interactions with the building envelope and building systems [9].

A great potential for energy savings would be achieved if occupants improved their awareness about their behavior and its impact on building energy use [15]. There are various approaches presented in the literature to enhance awareness and knowledge among building occupants. The use of digital tools such as smart phones or other smart displays is often showcased, as they are proven to be successful when providing awareness through feedback and gamification features [16].

The conventional building landscape is transforming into a smart building landscape due to the rapidly increasing amount of sensor data enabled by Industry 4.0 and digitalization. It is therefore relevant to address how smart building systems will affect buildings and understand how the user interface changes. In buildings with conventional building systems, the building occupant can interact with the building systems, for example, by manually turning on or off lighting or changing thermostat set points. These interactions are often mentioned in the literature as problematic as they could lead to an increase in building energy use.

The literature is, to a large extent, focused on office and residential buildings with conventional building systems. It is therefore relevant to enhance our understanding of building occupant behavior in commercial buildings equipped with smart building systems.

1.1. Campus Buildings

There are several reasons why campus buildings are important environments to study in. These buildings host a variety of energy-demanding learning activities, such as laboratories, computer rooms and server rooms. High energy use is also related to inefficient use of indoor space, scheduling and variation in occupancy densities [17]. In addition, HVAC systems and lighting are depicted as the main energy consumers [18], even though it should be noted that the use of LED lights significantly reduces building energy use. On a global scale, as the demand and number of enrollments in higher education increase [19], the energy use for commercial buildings, including campus buildings, is estimated to increase by 1.6 percent per year from 2012 to 2040 [20].

Campus areas can be seen as small-scale cities with their diversity of multifunctional buildings, both residential and commercial. People live, work and study at campuses, resulting in a vivid atmosphere both daytime and nighttime [21,22]. Therefore, these areas are important to target in the sense that they could be seen as real-world laboratories on a large scale where innovative ideas could be explored [12]. Innovative research can also be tested in a controlled environment in order to later be implemented on a larger scale [23]. Furthermore, campus areas create a densely populated zone with energy-intensive properties; hence, engaging in energy-saving activities provides students with an opportunity to develop environmentally friendly behaviors that they will bring into their professional lives [11,24,25,26].

It is also stated in the literature that people working, visiting, living and studying in campus areas are generally not concerned about how their behavior affects the energy use of the campus building. A change in behavior is thus complex due to the absence of motivation and interest [27]. Campus buildings also have a unique setting of energy-demanding activities, which, together with the resulting behaviors, call for attention [28]. Furthermore, it is found that occupancy patterns change radically over time [29], which adds complexity to forecasting building energy use.

The role of campus areas as a platform for sharing best practices and as responsible societal actors has grown [23]. This can be done by the means of living labs, that is developed for transforming traditional higher education towards the next generation of learning, including meeting societal and business-driven demands [23,30,31]. Several campuses have also created their own sustainability agendas and joined networks, such as the International Sustainable Campus Network (ISCN). The ISCN holds particularly one principle relevant for the scope of this paper: that its campuses have an opportunity as a living laboratory for outreach to industry, government and organized civil society to come together about economic, environmental and social matters [22].

1.2. Research Gap

There are fewer studies on energy use in buildings with shared spaces. According to a literature review made by Delzendeh et al. [15] around 75 percent of the 100 reviewed articles dealt with residential or office buildings, which is in line with the findings of this work, where several of the reviewed materials stated that there are fewer research contributions that study higher educational buildings or shared spaces in general [32,33,34,35]. In addition, the challenge of attaining energy-aware behavior amongst campus building occupants calls for further research about building occupant behavior in campus settings.

Given this background, it can be stated that the relationship between smart building systems and occupant behavior is multidimensional and complex and many parameters are yet to be understood. Therefore, the objective of this literature review is to target this research gap, critically review selected literature, present current understandings and provide novel expertise by highlighting existing gaps and proposing areas for future research. This will be useful for building owners, smart system providers, smart system designers and other key stakeholders in this field.

2. Methodology

This study is a semi-systematic literature review with a meta-analysis approach focusing on studies targeting building occupant behavior, building systems and energy use in educational buildings. This study also includes an approach to integrative literature review, as the aim is to critically review selected research, contribute new knowledge in a multidisciplinary field, identify research gaps and provide suggestions for relevant future research [36]. This research process is reproducible and .csv files of the two database search results will be provided upon request from any of the authors. The research process is illustrated in the flowchart in Figure 1.

2.1. Conceptualization

Conceptualization was made by defining the problem within the boundaries of campus buildings, smart building systems, energy use and building occupant behavior. Six research questions (RQ1–RQ6) were formulated in order to structure the literature review and to provide relevant findings from the selected literature and frame the discussion. It should be noted that the process of defining the research questions was an iterative process that evolved along with the findings from the selected literature. The final research questions were defined accordingly:

• RQ1: Which methods and tools are presented in the literature?
• RQ2: Is there a connection between behavior and the level of smartness of the building systems?
• RQ3: Are there any synergies between potential energy savings, building occupant behavior and smart building systems?
• RQ4: What are the proposed energy-saving potentials in simulations and experimental setups, respectively?
• RQ5: Are there any success stories resulting in the long-term effects of maintaining energy-aware behavior?
• RQ6: What are the reported obstacles to achieving energy-aware behavior?

2.2. Paper Selection Process

The literature review is based on peer-reviewed conference papers or scientific research published in academic journals. The Scopus and Web of Science databases were selected due to their extensive databases of peer-reviewed research. The Scopus database is considered to be the most effective database for literature reviews [37], and the Web of Science is the world´s oldest and leading scientific database of multidisciplinary journals [38].

The keywords were combined accordingly: ((Universit* OR “Higher educational” OR Campus) AND buildings AND (Occupant OR User) AND Energy AND (Behavior OR Behavior)). The same keyword combination was used in Web of Science and Scopus, resulting in 283 results (Scopus) and 248 results (Web of Science). The semantics were combined accordingly: “occupant OR user” and “Universit* OR higher educational OR campus”, were combined to enhance the search result. “Behavior OR behaviour”, was inserted to capture both spellings.

The Boolean operators “AND” and “OR” were used for the search query, and the search field “Topic” was selected to narrow the search results. There was no time filter applied, and all reviewed material was published between 2006 and 2022. The search process was refined by selecting the “Topic” field in Web of Science and the search within: “Article title, Abstract, Keywords” in Scopus.

A total of 531 articles were retrieved by combining the keywords, of which 170 were duplicates. A first screening looked for relevant terms or words in either the title or the abstract, resulting in 120 articles. A second screening considered the full length of the papers to identify relevant reasoning regarding any of the research questions, resulting in 84 articles. This process followed a literature selection process according to the Prisma framework [39], illustrated in Figure 1. It should also be noted that this literature review does not aim to include every scientific paper ever published in this field.

2.3. Data Analysis

The research questions were systematically treated, one by one, while going through the identified references. The selected papers were thoroughly analyzed for each research question. The results and conclusion sections of the papers were first analyzed to find any relevance to any of the research questions. If relevant, the full length of the paper was analyzed. According to Figure 1, this process was iterative, and therefore, the formulation of the research questions, including the literature analysis, was an evolving process.

2.4. Limitations

This literature review is limited to campus buildings, with a focus on buildings of educational character or mixed-use buildings with offices, classrooms and laboratories. Office buildings, dormitories or any other on-campus housing are excluded. The selected combination of keywords also frames a limitation, as there could be relevant literature outside the search results of these pre-defined keywords. Furthermore, it is recognized that targeting campus buildings involves a homogenous group, meaning people generally in the same age range with the same level of education. Any aspect of water use is excluded due to the relatively low water use from a building occupant perspective in educational buildings.

2.5. Results

The main contribution of this paper is based on the findings of the six research questions discussed under Section 3. Section 3.2.4 summarizes the energy-saving potentials provided by the reviewed literature. Also, a conceptual framework is proposed under Section 3.3, representing a novelty by mapping human factors in relation to building energy use. Furthermore, based on the identified research gaps in the literature, suggestions for future research will be proposed.

3. Results and Discussion

This section presents and discusses key findings from the literature review. First, the general findings of this literature review will be presented. Secondly, the findings based on the six predefined research questions will be discussed under each subheading. Thirdly, based on findings from the literature, the conceptual framework will be presented.

3.1. General Findings

The reviewed literature consists of multidisciplinary research dealing with social psychology, innovative technologies and building physics. Figure 2 represents the number of articles per year of publication. Given this result, research in the field of occupant behavior and its impact on building energy use has significantly increased since 2014. This increasing interest in research about occupant behavior, energy and buildings could be explained by the accelerated demand for providing energy-efficient systems and solutions to the built environment.

It is recognized in the literature that building occupant behavior is a key driver for a building´s energy use [32]. Therefore, research is trying to understand the implications of user behavior on building energy use. According to the findings of Pujani et al. [40], changes in behavior could result in energy savings. However, it requires knowledge about energy savings, which demands information and education.

Innovative technology such as the Internet of Things (IoT), including sensor networks, is emerging within the field of building technology. A few papers discussed the implementation of IoT and smart sensor networks to learn more about occupant behavior based on the analysis of sensor data. Sensor data enable new knowledge about how campus areas and buildings are utilized by campus users and could also provide input for energy-management instruments [16]. One of the key results in the study of Pujani et al. [40] was that the use of automatic sensor and control technology is the ultimate option to achieve long-term energy savings.

Digital Twin technology is an emerging technology that allow 3D visualization of buildings, where real-time sensor data could be combined and analyzed in order to identify and predict the building’s condition. Two papers presented digital twins as a tool for energy savings. In the first paper, Seo and Yun [41] developed a digital twin to assess the use of light in an energy-saving way. Their digital twin mirrored the buildings lighting system operations. Based on occupancy patterns as a result of scheduling, different scenarios were established that enabled energy-saving strategies. The results indicated that there would be great energy savings from lighting by using the digital twin to quantify and assess energy-saving potential. The second paper, Tagliabue et al. [42], developed a digital twin for the main purpose of assessing green building rating systems. The digital twin concept aimed at communicating sensor data to building users with the purpose of encouraging sustainable use and operation of the building.

Terminology and Individual Drivers for Thermal Comfort

Understanding the drivers behind building occupant behavior is complex and requires knowledge about the building occupants and the building itself, including environmental factors such as climate and building orientation. In this context, building occupant behavior is often termed energy aware, energy unaware or energy wasteful. This terminology will also be utilized by this study, as it is already an established phrase. These behaviors tend to have a strong correlation with the thermal comfort and wellbeing of the building occupant. Metabolism, layers of clothing, activity and where the building occupant is positioned in a room are examples of factors that affect the sense of thermal comfort.

According to several papers, examples of behavior that affects total building energy use often arise by influencing the indoor temperature through changing heating or cooling set points. Adjusting thermostats to a higher value than necessary instead of wearing an additional layer of clothing is one example [43]. It should also be noted that adjusting the level of clothing is an individual response to perceived thermal comfort and should therefore be considered individually. Hence, it is likely that opinions about temperature will vary within a group of people, as someone will perceive the indoor climate as too cold at the same time as someone else will perceive it as too warm. A typical classroom setting is a common area with a group of people. It is therefore challenging to achieve a thermal comfort that suits everyone since this is subjective and might differ between individuals or groups [44].

3.2. The Research Questions

3.2.1. RQ1: Which Methods and Tools Are Presented in the Literature?

The literature presented a number of methods and approaches to estimate, understand and predict the impact on building energy use as a result of occupant behaviors. As stated in the introduction, dealing with building occupant behavior is a complex matter, which most likely contributes to research presenting diverse results. It seems like there is no ultimate research method within this multidisciplinary field and most likely, appropriate research methods should be thoroughly selected depending on the context of building type, activities and surrounding environmental factors. The various approaches to dealing with occupant behavior are relevant to acknowledge since they illustrate the complexity of studying energy use, behaviors and buildings.

One method often used is computer-based modeling. Occupant behavior models can identify driving forces that explain energy-related behavior as well as understand connections between usage and energy demand [5]. Simulating occupant behavior has limitations since it is difficult to predict human behavior due to the complexity of their nature as human beings. It should not be expected that simulation-based research will be able to provide any exact predictions of building energy performance [45,46]. It should also be noted that in order to improve building energy use, it is important to understand occupant behavior in real buildings [13]. In line with this argument, Azar et al. [47] developed a model that simulated how people move and interact within an existing built environment, combined with building energy use and thermal comfort. The authors stated that this was an innovative approach because of the integration of a human-in-the-loop approach. Another example is Ding et al. [48], who developed a prediction method for commercial building electricity use where the building electricity use and occupancy were divided into a variable and a basic parameter. The variable parameter considered the electricity use due to the number of building occupants, and the basic parameter considered the building area. This model is aimed at accurately predicting building energy demand. Hence, the authors indicated an energy-saving potential due to efficient energy management by means of this model.

Questionnaire surveys are a research method often used to provide both qualitative and quantitative data. A total of 34 out of the 84 selected papers (40 percent) utilized questionnaires as an approach to gathering data and understanding energy-related behaviors. Surveys are used as a multidisciplinary approach to understanding the relations between building physics and social psychology to study the environmental triggers of building users. Soares et al. [12] performed a web-based survey with the intention of gaining knowledge about energy awareness and thereby mapping behaviors amongst both students and staff in campus areas. Results showed that campus users were relatively aware of behaviors concerning electrical devices, lighting, water and waste. In addition, Mataloto et al. [49] distributed a questionnaire amongst both students, lecturers and staff. The results indicated that there was a high level of awareness about sustainability matters; however, the respondents rated the level of consciousness amongst peers to be rather low. This is an interesting contradiction in the relationship between validity and interpretation. The validity of self-reporting through questionnaires could be argued about since several sources of error could arise due to questions being answered incorrectly. Sjöström et al. [50] argued that questionnaires are biased due to non-responses and inaccurate answers such as misinterpretation of the questions, delivering a false answer or simplifying the respondents beliefs, resulting in an incorrect answer.

Several papers used a variety of communication strategies to improve awareness and knowledge among building occupants, especially by means of digital tools. For example, Pujani et al. [40] developed an information dashboard to increase awareness among building occupants about various conditions related to building energy use, such as costs and carbon emissions. An unexpected outcome was presented in the work of Timm et al. [51], who also developed a dashboard to communicate building energy use in real time. Simultaneously, together with an energy behavior change campaign, the intention was to study if there would be any changes in attitudes or behaviors amongst building occupants. The results presented a decrease in building energy use; however, this was a result of interventions by the building operators. The dashboard presented aggregated information about building energy use; therefore, operators could easily detect any issues. The behavior of the building occupants seemed unchanged, as the dashboards were more beneficial for the building operators.

The literature presented a few communication strategies through the means of various user manuals and programs. Almeida et al. [52] suggested a building guideline manual, U-BOP (User´s Building Optimal Performance). The purpose was to provide instructions to building occupants about how to interact with a building to ensure energy efficiency. It is suggested that this manual be adapted to each individual building on-campus, meaning that the guidelines should consider all the various types of buildings and building systems. According to a survey conducted in this study, 81 percent of the building occupants had no access to any information about how their buildings operate. Beyond targeting the lack of building manuals, this manual also aimed to provide a tool to enhance occupant knowledge about how to interact with and operate building systems. Another approach to targeting building occupants’ awareness was elaborated by Marans and Edelstein [53], whose research aimed at understanding behaviors and attitudes towards saving energy amongst staff and students at an American campus. Building occupant data were gathered through a mixed-method approach, including interviews, focus groups and observations to collect input for online surveys. These analyses served as background material for an energy conservation program for campus buildings. This program was considered successful due to preliminary building data that indicated a decline in building energy use. The study could, however, not confirm that the energy reduction was due to this program and stated that more research was needed to validate the actual impact on energy use.

Another innovative concept is eco-feedback technology. This is a method that provides feedback to individuals or groups with the intention of reducing environmental impact. The literature seemed to agree that feedback approaches are successful methods of promoting energy awareness, as feedback is perceived to be an easy and effective method of teaching building occupants about these matters [54,55]. Gamification is a trend that is used in feedback studies to improve building occupants awareness about building energy use [16,56,57]. Gamification can be described as a trend that is new to this field, generally explained as bringing game mechanics into nongame environments. Gamification could also include reward mechanisms, for example, converting electricity savings into points that give some kind of reward. According to Ferreira et al. [17], there was a 40 percent decrease in electricity use from lighting during a two-month test period. Hence, gamification has been proven successful in the sense that it makes energy-related information more engaging and interesting while creating motivation for a change in behavior.

3.2.2. RQ2: Is There a Connection between Behavior and Level of Smartness of the Building Systems?

This research question is relevant to address since the advancement of technology affects the level of interaction among building occupants. Manually adjusting thermostats or switching on and off lighting are examples of building occupant interactions in buildings equipped with conventional building systems. In smart buildings, interactions between building systems and building occupants are limited, as, for example, smart HVAC systems and lighting are automatically steered based on sensor data.

None of the 81 selected studies provided sufficient details about the level of smartness of the building systems. In general, whether the building systems were smart or conventional could only be assumed depending on how interactions with the building systems were described. As an example, some studies categorized building occupants as energy unaware due to their reluctance to switch the lights off when they exit a classroom [57,58]. Since the building occupants actively switched on or off the lighting, it could be interpreted that these studies took place in buildings equipped with a conventional lighting system. Also, Aghniaey et al. [59] stated that their case study building is “fully digitalized”, but the authors never explained what it resembled in terms of building systems. Furthermore, it is stated in several papers that the case study buildings are equipped with HVAC systems, but it is not stated whether the HVAC system is equipped with sensors [59,60,61,62,63]. Some studies also state that the building occupants interact with the HVAC system [64], although it is not stated how this interaction occurs. When this information is left out, it lies in the reader’s assumptions to estimate the smartness of the building systems.

These missing descriptions reveal the complexity of dealing with building occupant behavior and that the relationships might not be fully understood. In studies dealing with energy-related behavior in buildings, the type of building systems should be presented as part of the methodology or theory section. This information would provide the reader with relevant background regarding building occupant behavior and its resulting impact on building energy use.

3.2.3. RQ3: Are There Any Synergies between Potential Energy Savings, Building Occupant Behavior and Smart Building Systems?

This research question was elaborated with the aim of understanding to what extent these linkages have been studied. It is evident that the level of interaction between building occupants and building systems is reflected in the total building energy use. In buildings equipped with smart building systems, this study argues that the level of interaction from the building occupants is limited due to improved automation, sensor-steered lighting, HVAC and design aspects such as smarter interfaces. Thereby, in buildings equipped with smart building systems, the potential to reduce building energy use also becomes less. As an example, in buildings equipped with smart HVAC systems that automatically control indoor climate, building occupants interact less with these systems compared to conventional systems. Therefore, as several papers discuss, reluctant behavior in regards to turning off the HVAC system when leaving a room or building should be less of a problem for a building with an HVAC system controlled in its presence [32,57]. Aghniaey et al. [59] approached this matter by stating that it is important to consider aspects of “smart occupants” and also “smart consumption behavior” when addressing energy optimization for in-door environments. This statement reflects on the smart building occupant, although it is not clear whether the authors referred to smart or conventional building systems.

Withanage et al. [33] found that smart building systems were requested by the building occupants in order to get automatic switch-off devices and automatic adjustments based on occupancy. The literature also confirmed that smart building systems have the potential to decrease building energy use while enabling an improved understanding of the occupant’s impact on building use by tracking and monitoring occupancy patterns through the use of sensor data [57,65]. For example, Pujani et al. [40] stated that the use of sensors and automatic controls would enable energy-saving opportunities. It was also recognized by Azar and Ansari [13] that in buildings equipped with sensor-steered lighting, the opportunity to save electricity from lights through targeting behavior is limited. As buildings become equipped with smart lighting systems based on presence or motion sensors, the impact of human behavior will be reduced when the habit of manually switching on or off lights is gradually removed. Information and communications technology (ICT) enables improved control, design and operation of building systems and is therefore important from an energy-saving point of view [22].

To conclude, no paper presented in detail any synergies between potential energy savings, building occupant behavior and smart building systems. As described above, only a few papers elaborated on the opportunities that digitalization brings when dealing with building occupant behavior, building technology and smart building energy use. Implications and future scenarios between these three important aspects of campus buildings were found to be missing. Since conventional building technology is gradually replaced with smart building technology, the building occupants’ interactions will change, and therefore, these aspects are vital to grasp in the sense of managing and operating smart campus buildings.

3.2.4. RQ4: What Are the Proposed Energy-Saving Potentials in Simulations and Experimental Setups, Respectively?

25 of the 84 articles presented energy-saving potentials through various interventions by means of simulations or experiments, including a wide range of energy-saving potentials by various interventions targeting interactions between building occupants. The results differed between simulations and experiments, as well as within each category. This research question aims at highlighting these findings and presenting possible explanations for these scattered results.

Figure 3 illustrates a summary of reported energy-saving potentials. On the left, results from simulations are presented, while the right side of the chart presents results from experiments. Each side is divided into light, HVAC, El and TBEU. Light represents energy-saving potential by targeting lighting systems, for example, installations of presence-steered lighting. The HVAC column presents results from interventions in relation to a building’s HVAC system. El is energy-saving potential from targeting any type of electrical appliance, for example, the use of stand-by functions. TBEU, total building energy use, is any type of intervention stated to improve total building energy use. Figure 3 follows a scheme from the work by Nguyen et al. [35].

The legend in Figure 3 describes the various results. The small figures represent results based on short-term studies ranging from 2 weeks to 4 months. The larger figures represent long-term studies ranging from 9 months to 3 years. The circle represents energy savings due to technology, meaning interventions based on technology-driven research. The triangle represents energy savings due to behavioral interventions, where the building occupant was the focal point for the result. The solid line circle/triangle represents measured, real results. The dotted circle/triangle represents potential savings not yet measured in a real building. It should also be noted that a few simulation-based studies achieved actual results in real buildings; therefore, the representation of solid line figures within simulations. Likewise, within experiments, there were results that presented potential energy savings that had been estimated but not yet confirmed in any real building.

Figure 3 represents a wide range of energy-saving potentials found in the literature. It should be noted that any relevant variables affecting the building should be considered [66]. While too few studies addressed climate zone, geographical location, building orientation, building envelope and season of the year [64], it should also be noted that further details about building systems are required for a full understanding of all the underlying factors behind energy savings.

A Hong Kong-based study [10] achieved a 23.5 percent reduction in the central air-conditioning system due to changes in user behavior. Hong Kong has a humid subtropical climate, which means that there is a high energy demand for cooling. Hence, targeting the air-conditioning system to achieve energy savings holds potential in this case due to the climate. Thus, the presented results in Figure 3 should merely be interpreted as indications of energy-saving potential.

Results obtained from simulations tended to be more optimistic compared to the results obtained from experiments, which are more modest. Five studies based on simulations presented the potential for at least 40 percent energy savings. One of these studies, performed by Awang et al. [67], claimed an energy-saving potential of 44 percent due to the replacement of conventional lighting systems (with manual switches) with lighting systems controlled by occupancy sensors.

A total of 12 papers presented experiments involving students, teachers and staff. A total of 13 papers presented results from simulation-based studies. Experiments performed in a real-world environment require resources such as time, effort, motivated people and funding. Experiments are also difficult to control, especially when dealing with human behavior that is complicated to predict. While simulations offer more complex research since it is possible to make assumptions. Using behavior as an input variable for energy performance simulations in educational buildings is particularly difficult due to the diverse behaviors and activities of the building occupants [65]. Klein et al. [68] highlighted the trade-off between thermal comfort and building energy use. They demonstrated in their study that they were able to reduce the energy use of the HVAC system by 17 percent while keeping 85 percent of the building’s occupants satisfied with their thermal comfort.

These scattered results presented in Figure 3 indicate that more research is needed to understand potential energy savings. To add to the complexity, some studies lack a detailed description of how energy savings are benchmarked. As energy savings could be calculated in various ways, it is unclear what this percentage represents. As an example, the simulated energy-saving potential of targeting building occupant behavior and HVAC systems is in the range of 10–45 percent. In comparison, experimental results presented a more modest result of 10–23.5 percent. Furthermore, as identified in RQ2, it is also challenging to understand if these results are based on buildings equipped with smart or conventional building systems.

3.2.5. RQ5: Are There Any Success Stories Resulting in Long-Term Effects of Maintaining Energy Aware Behavior?

None of the papers presented long-term results, and only a few acknowledged the importance of maintaining sustainable behavior. Boulton et al. [32] designed an energy challenge within a campus, including both students and staff. One goal was to learn whether there would be any lasting energy savings after the initiative was completed. One of their key findings was that variable data such as temperature, day length, campus population and energy retrofits must be thoroughly analyzed in order to realize energy-saving potential from behavioral aspects during a specific year. Due to these variables, energy data could be misleading. Therefore, it is important to understand if a decline in energy use depends on a seasonal factor, such as warmer temperatures, or if it is due to behavioral change.

Amaral et al. [69] highlighted the importance of long-term goals for various campus stakeholders; organizational, educational or operational. It was concluded that maintaining a desired long-term behavior is difficult, as users tend to forget to preserve awareness about energy-related behaviors. Another study by Timm and Deal [51] developed a dashboard that communicated building energy data. They concluded that achieving long-term behavior is related to a sense of control. In a commercial building, the occupants experience less control in general, which could be one reason why it seems difficult to uphold a desired energy-aware behavior. Yet, there were no successful long-term results presented in this study. However, the importance of understanding that user behavior, energy-saving norms, and attitudes must progress gradually to develop effective energy-saving strategies was highlighted.

According to Wai [70], a reminder is required to maintain a desired behavior. It is suggested that this could be solved by long-term activities that remind the users, for example, using feedback features. This is in line with the findings of Xu et al. [71], who found eco-feedback useful to promote long-term behavior at a community level. To conclude, it is challenging to maintain an energy-aware behavior amongst building occupants, and since the literature lacks results of long-term initiatives, knowledge is limited to the results of short-term studies.

3.2.6. RQ6: What Are the Reported Obstacles for Achieving Energy-Aware Behavior?

There are several identified obstacles that could be identified as bottlenecks for energy awareness among building occupants. The absence of knowledge and awareness is depicted as one of the most crucial obstacles to achieving energy efficiency [72]. Without awareness in place, people will be hindered from acting in an energy-saving manner. Hence, building occupants will not be able to adopt energy-saving behaviors without any concern about the significance of energy savings [70]. The literature presented several efforts for improving knowledge, and with the tools offered by digitalization, there are several potential solutions that could encourage long-lasting behavior. For example, providing building occupants with feedback via mobile apps, web apps or digital community boards is a method that is considered to have positive results.

Economic obstacles were also mentioned in the literature, as several papers stated that neglectfulness about energy-saving behavior can be related to economic incentives [26,73,74]. Electricity is perceived to be free to commercial building occupants. This is explained by the financial consequences of a behavior since the building occupant is charged for energy use in domestic living but not in commercial buildings [17]. This could cause wasteful behavior, leading to an increase in building energy use, since it has no personal consequences in terms of costs. Kaplowitz et al. [75] interviewed lab personnel, students and other faculty members in order to identify barriers and incentives towards decreasing building energy use. The results revealed that the lab personnel did not manage the electricity bill or have any information about energy statistics and they were not particularly concerned about these matters. Also, due to budget constraints, the cheapest laboratory equipment was bought, which ignored energy-saving aspects. According to the authors, these findings seemed to be similar to previous research. There needs to be a correlation between laboratories, energy use and financial costs.

Another important obstacle to address is overconsumption. This is one of the most important obstacles to defeat since it leads to an unsustainable use of resources. It is also relevant to highlight this in the context of building occupant behavior since we spend over 90 percent of our time indoors [76]. Putting this fact in the context of a building occupant, as cooling, heating and lighting are the core energy-demanding sources on campuses, sustainable use must be pursued [26]. Pandey et al. [73] acknowledged overconsumption and the need for long-term strategies among larger groups of people. It was especially stated that social norms are crucial to target in order to achieve energy-aware behaviors.

Prafitasiwi and Rohman [7] identified, based on their literature review, a set of obstacles for reaching energy efficiency in campus buildings; “Lack of environmental concern, lack of awareness of energy saving, lack of policy and legislative measures, lack of communication from stakeholders, objections from vested interest groups, inadequate data and information, lack of incentive support and motivation, lack of social norms and lifestyles”. Although it is unclear what the authors mean by “objections from the vested interest groups”, most of these findings correspond to the findings in this literature review.

3.3. The Conceptual Framework

In order to develop relevant strategies for promoting energy-aware behavior, understanding the different characteristics behind occupant behavior and the resulting impact on building energy use is crucial. Based on findings from the literature review, a conceptual framework, illustrated in Figure 4, is proposed. The framework illuminates possible connections between three identified levels of mechanisms in the system relationship between building occupants and building energy use. As these connections are complex and challenging to describe [77], this framework aims to provide a structure to improve this understanding.

The framework should be interpreted as follows: The arrows represent possible connections between the three levels of assessment. Firstly, possible human factors that could explain the building occupant’s behavior are identified. Secondly, an implication of the level of awareness is coupled, and thirdly, the level of expected impact on building energy use can be defined.

The conceptual framework aims to provide a tool for the evaluation of human factors, their linked implications and their resulting impact on building energy use. Thereby, the framework categorizes an expected level of impact on building energy use based on human factors. The framework also targets an identified research gap by exemplifying key behavioral factors of building energy use [12].

An understanding of which human factors prompt building occupants’ behavior would also enable building practitioners to differentiate between these given impacts and ultimately prioritize interventions towards improving building energy use. Suitable interventions are not in the scope of this paper; hence, more research is needed in order to suggest applicable measures for how to fully use this framework. It should also be noted that this framework was developed with regards to commercial buildings.

3.3.1. Human Factors

The framework is based on ten human factors: Level of knowledge, values, gender, age, beliefs, social norms, habits, control, economic incentives and culture. While several of these are sociodemographic factors, the literature describes them as justifications for energy-related behavior. These human factors could be identified through survey questionnaires or interviews with the building occupants. It should be noted that there might be additional relevant human factors that could be added to this framework. However, as stated above, this framework is limited to human factors addressed by the literature.

The level of knowledge is claimed to be one human factor behind a specific behavior. Building occupants who behave in an energy-aware manner are well educated about energy savings and environmental matters. Several papers stated that the more educated building occupants are about building energy use, the more awareness will follow [45,58,73]. In contrast, energy wasteful behavior is connected to a lower education on environmental issues [7,78].

Values and beliefs are two human factors that are argued to influence the way building occupants act. It is recognized in the work by Pandey et al. [73] that in multicultural settings, values and beliefs tend to differ. These aspects are relevant to consider since campus areas usually hold a mixture of nationalities. Beliefs were also treated by Azar et al. [79], who studied the connection between energy-aware behavior and beliefs. They pursued a survey where the respondents answered questions about their energy-related behavior, and the results demonstrated that beliefs were one of the main factors behind energy-saving behavior.

A few papers depicted cultural aspects as explanations for building occupants’ energy-related behavior. The authors of the IPCC fifth assessment report [80] found it evident that buildings with similar facilities and the same number of people could differ by a factor of 2–10 regarding energy use. The cultural aspects are especially important to notice since the campus building hosts occupants with diverse cultural backgrounds.

When dealing with energy-related behavior in commercial buildings, social norms are also important factors to consider. According to McDonald and Crandall [81] (p. 147), “A social norm is an expectation about appropriate behavior that occurs in a group context”. Also, Almeida et al. [52] argued that norms have an impact on behavior since energy-related behaviors are influenced by perceptions about what other people expect about behaviors, resulting in choices being based on perceptions of moral obligations. Furthermore, Azar et al. [79] depicted social influence as a key factor in energy-saving behavior. Neighbors and peers could positively influence energy-related behavior [51,52].

Habits might also affect building occupant behavior. As an example, the act of deciding when to switch on or off lighting is under habitual control and is thereby linked to building occupant behavior [82]. Cantù [83] addressed habits, stating that bad habits could lead to an increase in building energy use because the way people interact has a great impact on building energy use. Furthermore, Jowkar et al. [44] indicated that habits are an explanation for thermal comfort preferences, as it is evident that people´s comfort temperatures correspond to their exposure to a certain thermal environment. But what is foremost important to acknowledge is that habits can also lead to significant improvements in building energy use and are therefore important for understanding the drivers behind building energy use [84].

A sense of control in an environment is, from a psychological perspective, important for the building occupants’ perceptions of their thermal comfort [44,73]. According to survey results from Azar and Ansari [13], the perceived lack of control over building systems in commercial buildings, compared to domestic buildings, could be a reason for the perceived lack of a mandate to save energy. Hence, there is a willingness, but not the right fundamentals.

Demographics, especially gender and age, are studied in a few papers. Aghniaey et al. [43] found in their study that women generally experienced the thermal indoor environment as colder than men, even if they wore somewhat more clothing. Also, Wang et al. [58] found that men worried less about saving energy, and their comfort preferences tended to be higher, compared to women.

The lack of economic incentives is depicted as a factor in energy wasteful behavior, as building occupants in educational buildings are not charged for energy use. Hence, wasteful behavior is reflected in the belief that energy use comes with no cost [26,73,74].

3.3.2. The Implication

It is argued in the literature that human factors lead to behavior that is either aware, unaware or wasteful. Once the human factor amongst the building occupants is determined, the second step will be to determine the implication of aware behavior, unaware behavior, aware-wasteful behavior or unaware-wasteful behavior. Aware-wasteful behavior was included to capture negligent behavior, although the consequences of the behavior are understood. Unaware-wasteful behavior captures the behavior of unintentionally being wasteful. The implications could be determined by analyzing data from occupancy sensors in combination with building energy use and information from scheduling and room booking systems.

3.3.3. The Expected Impact of User Behavior on Building Energy Use

The third step would be to determine the expected level of impact: high, variable, low or uncertain. This could be done by analyzing building energy data and benchmarking it against forecasted building energy use. Normalized factors for the particular building of study should also be included, taking into account, for example, local conditions such as climate. Variable impact and uncertain impact were not mentioned in the literature; the majority of papers either pointed to a high or low impact. As building occupant behavior realistically ought to imply a more diversified impact than either high or low, this framework also considers variable and uncertain impacts. A future development of this framework could quantitatively define each impact with a set of parameters. That would enhance our understanding of the actual impact on building energy use. Furthermore, it should be noted that there could be additional aspects to consider at all three levels of the framework.

4. Conclusions

This literature review targeted research about building occupant behavior and its resulting impact on building energy use. The type of building selected for the review was campus buildings with an educational character, such as laboratories and classrooms. There are several reasons why this specific building type is important to target. Firstly, the demand for campus buildings is expected to increase, as is the energy use. Secondly, campuses are where the future work force is educated to be responsible citizens and where they should learn about how to maintain long-term energy-aware behaviors in order to achieve a sustainable built environment. Thirdly, the literature confirms that there is a research gap in sustainable energy use for higher educational buildings, whereas residential and office buildings have been studied to a greater extent. Lastly, campus buildings are energy-demanding due to facilities such as labs, computer rooms and use halls and their specific occupancy patterns. This is all crucial for the pathway to achieving a climate-neutral building stock.

The literature agrees that optimization of total building energy use and occupant behavior can be conflicting. The underlying reason is that building occupants require thermal comfort and with no or very little knowledge, they interfere with the building envelope and the building systems. This interference leads to an increase in building energy use as a result of an extra load for the operation of building systems. Thus, targeting building occupant behavior has the potential to improve building energy use. Questionnaire surveys were a commonly used approach to gaining knowledge about the energy awareness of building occupants. Enlightening building occupants about their resulting impact on building energy use has been shown to have positive results and this could be done through information campaigns and energy-saving competitions. In addition, providing education about how to interact with the building and its systems would give the building occupants better knowledge about their resulting impact. It should be noted that such intervention requires resources from the building management team in terms of time and planning. It also requires continuous repetition as campus users interchange every semester.

A conceptual framework, presented in Figure 4, is proposed. It captures the relationship between identified human factors, energy-related behavior and the resulting impact on building energy use. The conceptual framework could be utilized as an evaluation tool for mapping human factors in relation to building energy use. The framework targets a research gap in understanding and evaluating how human factors relate to behaviors and energy use. The framework could be used by building management in order to gain knowledge about the building occupants and their drivers behind energy-related behaviors. Once this knowledge is acquired, the building management will be able to take appropriate actions to mitigate energy-wasting behaviors.

Conventional building technologies are continuously being replaced with smart building technologies as buildings are renovated or new buildings are constructed. Simultaneously, the role of the building occupants is likely to change as interactions are linked to the smartness and design of the building systems, ultimately impacting the resulting building energy use. It was found that no study investigated the relationship between building occupants and smart building systems. This observation is significant for future research to consider when studying building occupant behavior and its impact on building total energy use. An enhanced understanding of how smart building systems are linked with energy-related behavior will enable a better foundation for developing appropriate tools and methods for promoting energy awareness with current technologies in use.

Based on the key findings from this literature review, areas for future research are proposed:

• The lack of addressing the perspective of smart building systems when studying building occupant behavior and the resulting impact on building energy use in commercial buildings represents a research gap that should be addressed by future research.
• The reported energy savings presented in Figure 3 represent scattered research results and expose a challenge in the complexity of studying energy savings in commercial buildings. Future research should therefore invest in appropriate guidelines and methods for evaluating potential energy savings.
• The conceptual framework presented in Figure 4 should be tested in a real commercial building to validate the tool and further develop the concept in order to fill the research gap about human factors and their resulting impact on building energy use.
• The literature presents a lack of long-term results for maintaining energy awareness among building occupants. Campus buildings also face the challenges of a continuous exchange of building occupants with diverse backgrounds. Future research should therefore aim at long-term studies that especially consider the complex and dynamic challenge of achieving lasting energy-aware behaviors in campus environments.