Maximizing Energy Performance of University Campus Buildings through BIM Software and Multicriteria Optimization Methods

Abstract

University buildings have high energy requirements due to their size, numerous users, and activities, which considerably contribute to environmental contamination. Implementing energy-saving solutions in these structures has a favorable influence on the economics and the conservation of energy resources. A higher education building’s energy behavior can be simulated using software to identify the optimal strategies that result in energy savings. In this research, Autodesk Revit, Autodesk Insight, and Green Building Studio are among the programs utilized to examine the energy efficiency of the university building in four European cities. Following the development of several energy-saving scenarios for the building, the offered solutions are evaluated based on their annual energy consumption, energy costs, and ${\mathrm{CO}}_2$ emissions. Finally, multicriteria analysis techniques such as the AHP and PROMETHEE are applied to choose the best scenario for each instance. The study’s findings indicate that the ASHRAE Terminal Package Heat Pump scenario performed well in all of the cities examined, reducing yearly energy usage by 43.75% in Wien and annual energy costs by 47.31% in Mallorca. In comparison, the scenario utilizing a high-efficiency VAV system with a gas boiler and chiller came in last in all situations, resulting in a decrease of 12.67% in Mallorca’s annual energy usage and a reduction of 17.57% in Palermo’s annual energy expenses.

1. Introduction

The effect of complex and diverse factors and activities makes the energy research of higher education facilities challenging. Mohd Shukri et al. [1] declared that buildings utilize almost 36% of total energy consumed and produce 30% of total carbon dioxide (${\mathrm{CO}}_2$) emissions. According to Xing et al. [2], the energy consumption of American universities amounts to 13% of overall building energy consumption, whereas French institutions consume 38% of public facilities. Alshibani [3] claimed that university buildings account for around 13% of energy usage in Saudi Arabia. As Kim et al. [4] reported, in Korea, the university’s energy consumption accounts for 13.8% of the building sector. In universities, it is possible to face energy waste in different spaces such as classrooms, offices, laboratories, computer rooms, or dormitories. Applying organizational, technical, and energy efficiency strategies might significantly lessen an educational building’s impact on the depletion of natural resources and the pollution of the environment. Large-scale renovations of university buildings provide an opportunity for cost-effective measures to improve their efficiency [5]. With appropriate technical and cost-effective technologies, it is possible to improve the energy efficiency of existing buildings significantly, with significant environmental, economic, and social benefits [6].

In recent decades, researchers have devoted themselves to improving the energy efficiency of buildings through various technologies and strategies. Simulating a building’s energy behavior with software is essential for various reasons, such as cost reduction, promoting alternative energy forms, and predicting and optimizing energy consumption. Extensive literature attempts to capture economically viable approaches and scenarios to increase the energy efficiency of educational buildings at cost-optimal levels. Li et al. [7] studied 24 campus buildings to investigate their energy efficiency and used the percentages of floor areas of different spaces to perform internal benchmarking. This study demonstrated that the suggested approach can offer some guidance to facility management to evaluate and rank the energy-related concerns of the building. The energy efficiency of buildings used primarily as laboratories and the discovery of inefficient structures that were not immediately apparent in the initial EUI comparison are two examples of novel findings. In order to identify opportunities for electrical energy efficiency improvements in buildings and specify energy management strategies related to electricity use, Batlle et al. [8] took into account the recommendations of ISO 50001:2011 and ISO 50006:2014 for inducting energy benchmarks and energy performance metrics in educational buildings. In order to do so, a variety of variables influencing energy consumption in campus buildings, both controllable (operating hours, construction materials, number of people, equipment, and location) and uncontrollable (weather conditions) were analyzed and taken into account. The findings indicated that the campus’s potential yearly savings in electric energy consumption may be roughly 9.6%, resulting in a 20.3 t ${\mathrm{CO}}_2$ eq emissions reduction. However, the values of the energy usage parameters examined in this paper were based on limited data collection due to the risk of personal information leaking. Kolokotsa et al. [9] presented an effective web-based energy management system for universities that inspects the energy load, conducts energy analyses, controls campus buildings in an energy-efficient manner, and communicates with each user through emails and forms. The online survey of building users demonstrated a significant increase in tenant satisfaction and an energy efficiency of more than 20%. Analyzing the energy consumption of several LEED-certified and non-LEED-certified University U.S. buildings to identify potential sustainable energy policies, Agdas et al. [10] concluded that no findings support the conclusion that LEED-certified buildings use less energy than non-certified buildings. Chung et al. [11] presented the results of the energy analysis of campus buildings in Korea after the application of energy-saving policies. The interventions included window replacements, strengthening of the thermal insulation adequacy of the walls, and the addition of automation systems. The energy evaluations of the examined buildings revealed a possibility for saving energy ranging from 6% to 30%, depending on the type of space. Baniasadi et al. [12] examined the installation of photovoltaic rooftop systems in residential buildings with electrical and thermal energy storage systems to reduce electricity and life cycle costs. For that purpose, a prediction model that controls the energy flow of the storage system was developed. The power flow of the battery can serve all loads during peak-load hours. By managing the heat pump’s operation, it is possible to regulate the water flow in the storage tank to satisfy the required thermal energy needs. The previous actions save more than 80% on yearly power costs and 42% on life cycle costs. Oyedepo et al. [13] utilized an energy modeling application to evaluate Covenant University’s energy consumption and make energy-saving recommendations for 18 selected buildings. While yearly energy savings with a payback time of fewer than six years are achievable, Ref. [13] does not address the issue of replacing old, high-energy-consumption air conditioners with newer, lower-power-consumption models. Litardo et al. [14] investigated the energy efficiency of campus buildings, looking at the prospects for building energy savings using different passive and active measures, as well as on-site power production using a grid-connected solar PV system. However, the proposed model is based on tropical conditions and may not be relevant to communities with other climates.

Building information modeling (BIM) tools have been used in the last few years to analyze buildings’ energy behavior. Building information modeling is a trustworthy digital representation of a building that may be used for design decisions, high-quality construction documentation, construction planning, performance forecasts, and cost estimations [15]. The BIM delivers up-to-date and reliable project design, cost information, timelines, energy analysis, structural design, and other information needed for building projects [15]. The capacity to coordinate changes and maintain consistency is a crucial feature of BIM software. Laine et al. [16] examined the advantages of using BIM software in the energy analysis of buildings in contrast to traditionally zone-based models. They concluded that BIM provides more effective data entry and reuse of existing data, and the ability to employ dynamic energy modeling, and allows for ongoing verification of energy performance throughout the structure’s life cycle. Using BIM software, Venkataraman et al. [17] performed an energy study on a two-story residential building in six different climatic zones in India. Jangalve et al. [18] made an energy analysis of a residential building using BIM in its current state without providing any energy-saving scenarios. Tahmasebinia et al. [19] implemented a BIM energy analysis of a college building in Sydney to ascertain how the building’s shape affected energy consumption and predict future energy use. However, the authors did not study how using renewable energy sources (solar energy, wind energy, biomass, etc.) might result in energy savings or how alternative HVAC systems affect energy use. Del Ama Gonzalo et al. [20] investigated the energy behavior of a single-family house in Almeria, Spain, suggesting as optimization possibilities an increase in thermal resistance in walls and roof, the use of glazing in windows, and the inclusion of some photovoltaic solar panels. Nevertheless, the authors did not examine how different HVAC systems affect the energy consumption of this type of building.

Energy optimization of educational buildings is necessary due to their extensive energy consumption. The importance of conducting this research is highlighting viable tools for handling complex and multidimensional challenges, such as choosing between many energy optimization scenarios. For this reason, multicriteria techniques such as AHP and PROMETHEE were presented and analyzed thoroughly. Throughout this investigation, all processes were studied, and software solutions were offered for energy investigators to create the best energy optimization intervention while taking into account economic, technological, social, and environmental variables. In contrast to previous research, one distinctive feature of this paper is the examination of energy upgrades for a typical university building in not one, but four European locations with various climate conditions. The energy behavior of the university building was simulated with BIM software, and the results were analyzed using Multicriteria Analysis Methods [21,22]. In this study, the software Autodesk Revit, Autodesk Insight, and Green Building Studio is used to investigate the energy performance of the educational building.

2. Methodology

The purpose of this research is the energy evaluation of a university building in four European cities with different climate conditions while proposing various energy-saving scenarios. The interventions studied concerned the strengthening of the thermal insulation adequacy of the building, the installation of photovoltaic systems, and the replacement of heating and cooling systems. In order to determine the best combinations of interventions, it was considered necessary to simulate the building with some software to evaluate its energy behavior in different conditions. The energy analysis findings of the interventions evaluated for all four European cities are employed, and multicriteria analysis methods are applied to approach the optimal interventions for each scenario.

The stages followed for the implementation of this study are presented below:

Stage 1—Determination of the thermal characteristics of the building’s envelope, the spaces, and the location.

Stage 2—Cooling and heating load calculation.

Stage 3—Alternative, energy-saving scenario proposition.

Stage 4—Calculation of annual energy consumption (electricity and fuel), energy costs, and ${\mathrm{CO}}_2$ emissions of all scenarios.

Stage 5—Use of multicriteria analysis methods (AHP and PROMETHEE methods) in order to determine the best scenario for each case.

Autodesk Revit software is used to analyze the thermal and cooling loads of the university building, but also to digitally represent all of a building’s physical and functional elements. Following, the simulation of the energy behavior of the University Building is carried out using Autodesk Insight and the Green Building Studio software. All energy optimization scenarios’ annual energy consumption, costs, and ${\mathrm{CO}}_2$ emissions are then compared. Finally, the scenarios are ranked using the AHP and PROMETHEE methods to choose the best one for each case. Regarding the scenarios studied, the combinations examined are the following:

(a)

Use of ASHRAE Heat Pump for cooling and heating, adding photovoltaic panels, and strengthening the thermal insulation of the roof of the building;

(b)

Use of the High Efficiency Heat Pump for cooling and heating along with photovoltaic panels;

(c)

Use of the High Efficiency VAV system for cooling and heating along with photovoltaic panels;

(d)

Use of the ASHRAE Terminal Package Heat Pump system for cooling and heating along with photovoltaic panels.

2.1. Case Study

The building under study is a typical, mixed-use university building with a total area of 15,984 ${\mathrm{m}}^2$ and a total volume of 63,382 ${\mathrm{m}}^3$, consisting of classrooms, auditoriums, educational laboratories, offices, libraries, and public areas. It has six levels (basement, ground floor, and four storeys). The heights of all floors, basement, and ground floor, are equal to 4.05 m. In the current state of the building, a heat pump with a COP of 3.2 and an EER of 9.5 is utilized for space heating and cooling. The unit includes a heat recovery system with an integrated economizer and a hot water production system with an efficiency of 0.57. The following figure (Figure 1) presents the ground floor and first floor plans; a distinct color represents each space.

In

Table 1. Building Spaces and Areas (${\mathrm{m}}^2$).

Final Types	Intermediate Groups	$\mathbf{Area}\left({\mathbf{m}}^2\right)$	Area (%)
Laboratory	Laboratories	729	0.046
School Services	Classrooms	4066	0.255
	Offices	1942	0.121
	Auditoriums	984	0.062
Public Services	Lobbies	2635	0.165
	Food Services	128	0.008
	Library	208	0.013
	Stairway/Elevator	1131	0.071
Other	Mechanical/Electrical areas	197	0.012
	Warehouse	37	0.002
	Parking lots/Garage	3927	0.245

, the different spaces of the educational building and their areas are presented, while in

Table 2. Main characteristics of the University building.

Characteristics of the Building
Ground floor and floor ceilings (U = 0.306 W/(m²·K))	Glass fiber reinforced concrete (width: 0.03 m)
	Cement mortars (width: 0.01 m)
	Lightweight concrete (width: 0.05 m)
	Reinforced concrete (width: 0.1 m)
	Aluminum foil (width: 0.04 m)
	Cement lime mortar (width: 0.01 m)
Roof (U = 0.314 W/(m²·K))	Marble (width: 0.03 m)
	Cement mortars (width: 0.01 m)
	Lightweight concrete (width: 0.03 m)
	Reinforced concrete (width: 0.15 m)
	Expanded polystyrene in slabs (width: 0.10 m)
	Cement lime mortar (width: 0.01 m)
Exterior walls (U = 0.327 W/(m²·K))	Expanded polystyrene in slabs (width: 0.10 m)
	Triple coating (width: 0.01 m)
	Uncoated concrete (width: 0.2 m)
	Ceramic tiles (width: 0.02 m)
Windows (U = 2.921 W/(m²·K))	Large double-glazed windows
Doors (U = 2.5572 W/(m²·K))	Solid hardwood

, the structural characteristics of the building’s envelope and openings are given.

This study was selected to carry out the energy performance study of this building for four European cities (Mallorca, Palermo, Barletta-Andria-Trani, and Wien) (

Table 3. Annual Heating Degree Days (HDD) and Cooling Degree Days (CDD).

	HDD (18.3 °C)	CDD (26.7 °C)
Mallorca	1065	286
Palermo	1190	248
Barletta-Andria-Trani	1665	237
Wien	2717	95

).

2.2. Multicriteria Analysis

Multicriteria analysis refers to systematic procedures that solve complex and vital decision-making problems and offer an easy-to-understand approach to analytical logic. The selection between different energy optimization scenarios is a multifaceted decision-making process, including economic, technological, social, and environmental aspects. In this regard, the multicriteria analysis is a suitable method for combining and analyzing all viewpoints involved in the decision-making process by creating a link between all options and variables that impact the choice. It is critical to recognize that because there will be competing viewpoints and hypothetical solutions, the highest-ranking decision emerging from using MCDM approaches the best-negotiated solution rather than the best-stated answer [23]. In this paper, two multicriteria analysis methods are used: the Analytic Hierarchy Process (AHP) and the Preference Ranking Organization Method for Enrichment Evaluations (PROMETHEE) [24,25].

2.2.1. AHP

The Analytic Hierarchy Process is considered one of the most comprehensive methods for making multicriteria decisions. Analytic Hierarchy Process (AHP) was chosen in this study because of the socio-technical aspect of decision-making in energy systems. AHP is a prominent multicriteria decision-making model that allows analysis using quantitative and qualitative decision criteria. It can also manage several competing objectives while providing the option to deal with diverse stakeholder viewpoints. These criteria are very much in line with the nature of energy system decision-making. Furthermore, certain socio-technical factors are challenging to quantify, but AHP overcomes this barrier by allowing pair-wise comparisons [26].

AHP has been used in a variety of fields, including social science, politics, engineering, governance, and manufacturing. Serrano-Cinca et al. [27] investigated how AHP may respond to political circumstances in government administrations where the identification of public transportation routes and allocating public services is required. De Bin et al. [28] examined how AHP may provide solutions when organization, project structure, and resource allocation are needed in the business sector. Shatnawi et al. [29] used AHP to identify renewable energy solutions more suitable for Jordan’s climate.

AHP method is the only MDCM model that can calculate the consistency of respondents’ assessments, allows the formulation of the problem hierarchically, assigning a weight to each criterion and then creating the Pair-wise Comparison Matrix [30,31,32,33]. In order to create the Pair-wise Comparison Matrix, the following Scale of Relative Importance (

Table 4. Scale of Relative Importance [30].

Intensity of Importance on an Absolute Scale	Definition
1	Equal Importance
3	Moderate importance of one over another
5	Essential or strong importance
7	Very strong importance
9	Extreme importance
2,4,6,8	Intermediate values between the two adjacent judgments

) is used to attribute the intensity of the importance of one criterion to another.

(a)

Scale of Relative Importance

To create the Pair-wise Comparison Matrix, the following Scale of Relative Importance (Table 4) is used to assign the intensity of the importance of one criterion concerning another.

(b)

Pair-wise Comparison Matrix

Initially, a table is created that has the following format (Table 2), where {$c{r}_1, c{r}_2, \dots , c{r}_k$} denotes the set of different criteria studied while $x_{ij}$ and $\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$x_{ij}$}\right.$(i = 1, …, k, j = 1, …, k, i $\ne j$) denote the Relative Importance assigned to the criteria (where k = number of criteria) (

Table 5. Pair-wise Comparison Matrix [30].

	$\mathit{c}{\mathit{r}}_1$	$\mathit{c}{\mathit{r}}_2$	$\cdots$	$\mathit{c}{\mathit{r}}_{\mathit{k}}$
$\mathit{c}{\mathit{r}}_{\mathit{1}}$	1	$x_{12}$	$\cdots$	$x_{1k}$
$\mathit{c}{\mathit{r}}_{\mathit{2}}$	$\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$x_{12}$}\right.$	1	$\cdots$	$x_{2k}$
$\cdots$	$⋮$	$⋮$	$\ddots$	$⋮$
$\mathit{c}{\mathit{r}}_{\mathit{k}}$	$\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$x_{1k}$}\right.$	$\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$x_{2k}$}\right.$	$\cdots$	1

).

(c)

Normalized Pair-wise Comparison Matrix

To obtain the Normalized Pair-wise Comparison Table, each element of the Pair-wise Comparison Table is divided by the sum of the column to which it belongs. The weights of each criterion result from the sum of each row of the Normalized Pair-wise Comparison Table created.

(d)

Consistency Index

First, each element of the Pair-wise Comparison Matrix is multiplied by the weight of the corresponding criterion. These products are added for all columns, and then the weighted sum values are divided by the weight of each criterion. Thus, a ratio is obtained for each of them. Calculating the average of these ratios results in the value of ${\lambda }_{max}$.

The CI consistency index is then calculated using the following formula:

$$CI=\frac{{\lambda }_{max}-n}{n-1}$$

(1)

where n is the total number of criteria.

(e)

Random Consistency Index

Depending on the number (n) of criteria studied, the appropriate Random Consistency Index (RI) is selected (

Table 6. Random Consistency Index (RI) [30].

n	1	2	3	4	5	6	7	8	9	10
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49

).

(f)

Consistency Ratio

The Consistency Ratio is the comparison between the Consistency Index and the Random Consistency Index and is calculated through the following formula:

$$CR=\frac{CI}{RI}$$

(2)

If CR < 10%, then the consistency is acceptable.

2.2.2. PROMETHEE

PROMETHEE multicriteria analysis methods belong to the family of outranking methods and develop a methodological framework that allows performing pair-wise comparisons between the alternatives. PROMETHEE methods are considered some of the most widespread outranking methods due to their effectiveness in determining preferences between multiple decisions, their wide range of applications, and their ability to manage both quantitative and qualitative data [34,35]. Like all outranking methods, PROMETHEE methods involve preference relationships of strict preference, weak preference, and indifference. The two most frequently encountered PROMETHEE methods are:

• PROMETHEE I, which achieves a partial ranking of alternative scenarios;
• PROMETHEE II, which achieves a complete ranking of alternative scenarios.

Some of the least encountered PROMETHEE methods are:

• PROMETHEE III, which achieves an interval-based ranking;
• PROMETHEE IV, which is a generalization of PROMETHEE II in the case where the set of alternative actions is not finite;
• PROMETHEE V, which seeks to select a set of alternative actions under segmentation constraints (budget, performance, risk);
• PROMETHEE VI, which is a representation of the human brain.

The stages included in the PROMETHEE II method are presented below:

(a)

Determination of criteria weights

In order to determine the weights, either a multi-attribute method (such as AHP) must be selected or they must be set as equal [36]. Regardless of how the weights are calculated, the following limitation applies:

$${\displaystyle \sum }_{j=1}^n{w}_j=1$$

(3)

where $w_j$ are the weights of relative importance of the different criteria and $n$ is the total number of criteria.

(b)

Definition of preference function

As already mentioned, the operation of PROMETHEE methods is based on a pair-wise comparison between the alternative scenarios. A small discrepancy between the evaluations of two alternatives can mean weak preference or possibly indifference for the Decision Maker. At the same time, a greater deviation indicates a strict preference. The preference function has the following form:

$$P_j\left(a,b\right)=F_j\left[d_j\left(a,b\right)\right] \forall \left(a,b\right)\in A$$

(4)

where $d_j\left(a,b\right)$ is the difference of the evaluations a and b on each criterion.

(c)

Determination of the multicriteria preference index

The multicriteria preference index expresses the degree of total preference of one alternative a versus another b. Thus, for each pair of alternatives a and b, the index is computed as follows:

$$\pi \left(a,b\right)={\displaystyle \sum }_{j=1}^n{w}_j{P}_j\left(a,b\right)$$

(5)

(d)

Calculation of outranking flows

The outranking flows are distinguished into:

• Positive outranking flow: It reveals the superiority of the alternative a over all the other options and is calculated as follows:

  $${\phi }^{+}\left(a\right)=\frac{1}{n-1}\pi \left(a,x\right)$$

  (6)
• Negative outranking flow: It expresses how an alternative a is outranked by all the others and is measured as follows:

  $${\phi }^{-}\left(a\right)=\frac{1}{n-1}\pi \left(x,a\right)$$

  (7)
• Net outranking flow: It constructs an overall evaluation size of alternatives against all the remainders and is expressed through the following formula:

  $$\phi \left(a\right)={\phi }^{+}\left(a\right)-{\phi }^{-}\left(a\right)$$

  (8)

(e)

Complete ranking

The complete ranking of actions is done based on the net outranking flows. The higher the value of a net outranking flow, the higher the alternative ranks among the Decision Maker’s preferences.

2.3. Modeling Procedure

Building structural and functional components may be digitally represented using the BIM (Building Information Modeling) software Autodesk Revit. It includes the building attributes arranged in a database that the user can control, allowing for effective design and construction. Engineers may use it to analyze the building’s energy use and research heating, cooling, and ventilation. In addition to choosing the ideal temperature, humidity, operation times, and illumination levels, the user may also choose each space type. In particular, by using existing lists of materials, systems, and other parameters, researchers can more accurately analyze a building’s performance and better understand the energy model, ultimately evaluating which practices lead to a more energy-efficient building with minimal ${\mathrm{CO}}_2$ emissions.

Engineers may use the building volumes and details created in Autodesk Revit to undertake an energy evaluation of a building and find strategies to improve its energy use using the web tools Autodesk Insight and Green Building Studio. Multiple energy performance simulations of a building may be performed, and scenarios can be created based on the designer’s preferences. More precisely, the architect can alter the building’s orientation, the materials of the shell and frames, the operating hours, and the HVAC systems, and add solar systems enabling integrated energy analysis to be planned and presented for new and existing buildings. Figure 2 shows a photorealistic 3D rendering of the university building created in Autodesk Revit, while Figure 3, Figure 4, Figure 5 and Figure 6 show various views of the building.

3. Results

After the building spaces are created, and the thermal characteristics of the shell elements are defined, the desired conditions of temperature and humidity, lighting levels, and operating hours are selected. The space data are used to calculate the capacity of the building, as well as to perform energy analyses. In the table below (

Table 7. Cooling and heating loads and airflow summary table.

	Building in Mallorca	Building in Palermo	Building in Barletta-Andria-Trani	Building in Wien
Peak Cooling Total Load (W)	1,526,301	1,517,347	1,666,549	1,570,466
Peak Cooling Sensible Load (W)	1,049,856	938,823	1,086,542	897,502
Peak Cooling Latent Load (W)	476,445	578,525	580,007	672,964
Peak Heating Load (W)	732,388	782,253	855,132	1,312,741
Peak Cooling Airflow (L/s)	52,870	55,715	54,349	51,979
Peak Heating Airflow (L/s)	28,460	28,987	28,608	29,006

), the cooling and heating loads of the university building, as calculated by Revit, are presented.

Scenarios using air-cooled heat pumps with low, medium, and high-efficiency performance, scenarios using VAV systems with underfloor air distribution, and scenarios with gas boilers and chillers were examined to investigate the behavior of the university building in different HVAC systems (

Table 8. Scenarios examined in Autodesk Revit software.

	Heating	Cooling	Economizer	Water Production System	Photovoltaic Panels	Other Interventions
Existing building	ASHRAE 90.1-2010 minimum efficiency Heat Pump (COP = 3.2)	ASHRAE 90.1-2010 minimum efficiency Heat Pump (EER = 9.5)	70F	Eff = 0.57	-	-
Scenario 1	ASHRAE 90.1-2010 minimum efficiency Heat Pump (COP = 3.2)	ASHRAE 90.1-2010 minimum efficiency Heat Pump (EER = 9.5)	70F	Eff = 0.57	Cover 60% of the roof (18.6% efficiency)	Strengthening the roof’s thermal insulation with 10 cm of heat-insulating material.
Scenario 2	High-Efficiency Heat Pump (HSPF = 9.6)	High-Efficiency Heat Pump (SEER = 17.4)	70F	Eff = 0.57	Cover 60% of the roof (16% efficiency)	-
Scenario 3	High-Efficiency VAV, Underfloor Air Distribution, Gas Boiler (Eff = 0.95)	High-Efficiency VAV, Underfloor Air Distribution, Chiller (COP = 7.5)	70F	Eff = 0.57	Cover 75% of the roof (20.4% efficiency)	-
Scenario 4	ASHRAE Terminal Package Heat Pump (COP = 4.5)	ASHRAE Terminal Package Heat Pump (EER = 11.9)	70F	Eff = 0.57	Cover 60% of the roof (18.6% efficiency)	-

).

The diagrams below (Figure 7) present the total annual energy consumption, the annual cost of energy consumption, the annual consumption of electricity, and energy from fuels as they emerged from the software used. The analysis was made both for the present state of the building and for the energy optimization suggestions examined. Prices for electricity and natural gas for non-household users from the first half of 2022 were utilized for the cost analyses of the scenarios [37,38]. The cost of natural gas was 0.0729 €/kWh in Mallorca (Spain), 0.0734 €/kWh in Palermo and Barletta-Andria-Trani (Italy), and 0.0642 €/kWh in Wien (Austria). The cost of electricity varied from 0.1879 €/kWh in Mallorca (Spain), to 0.2525 €/kWh in Palermo and Barletta-Andria-Trani (Italy), and 0.1617 €/kWh in Wien (Austria).

The annual energy consumption of the university building in its current condition ranges from 88.5 kWh/${\mathrm{m}}^2$ in Mallorca, which has the lowest consumption, to 135 kWh/${\mathrm{m}}^2$ in Wien, which has the highest consumption. Scenario 4 delivered the most significant reduction in yearly energy usage in all cities studied: 35.65% in Mallorca, 40% in Palermo, 41.01% in Barletta-Andria-Trani, and 43.75% in Wien. Scenario 2 presented the lowest reduction in annual energy consumption in Palermo, Barletta-Andria-Trani, and Wien, achieving 11.67%, 10.07%, and 6.25% reduction, respectively. In comparison, Scenario 3 achieved the lowest reduction in annual energy consumption in Mallorca (12.67%).

Regarding annual energy costs, Scenario 2 resulted in the lowest annual cost reductions for Mallorca and Palermo (15.7% and 11.65%, respectively). Scenario 1, on the other hand, achieved the lowest reduction for Barletta-Andria-Trani and Wien (12.23% and 7.5%, respectively). Scenario 4 obtained the most significant decrease in yearly energy expenses in Mallorca and Palermo, saving 6.89 €/${\mathrm{m}}^2$ and 8.74 €/${\mathrm{m}}^2$, respectively. Scenario 2 achieved the most significant annual energy cost reduction in Barletta-Andria-Trani (45.07%), whereas Scenario 3 achieved the greatest annual energy cost reduction in Wien (53.14%).

In terms of annual electricity consumption, Scenario 3 achieved the largest reduction in all cases; 44.30% (31.22 kWh/${\mathrm{m}}^2$) in Mallorca, 46.24% (37.25 kWh/${\mathrm{m}}^2$) in Palermo, 66.09% (61.52 kWh/${\mathrm{m}}^2$) in Barletta-Andria-Trani, and 77.56% (81.21 kWh/${\mathrm{m}}^2$) in Wien. Scenario 2 obtained the lowest decrease in power usage in Mallorca, Palermo, and Barletta-Andria-Trani (15.7%, 11.65%, and 8.28%, respectively), while Scenario 1 achieved the lowest reduction of electricity consumption in Wien (5.62%).

Concerning annual fuel energy consumption, Scenario 4 produced the most significant decrease in Palermo, Barletta-Andria-Trani, and Wien (40.05%, 41.10%, and 45.09%, respectively), while Scenario 1 achieved the most considerable reduction in Mallorca (23.05%). Scenario 3 increased yearly fuel energy usage in all scenarios due to the use of gas boilers to satisfy heating requirements.

With respect to annual ${\mathrm{CO}}_2$ emissions, Scenario 1 achieved the most considerable reduction in Mallorca, Palermo, and Barletta-Andria-Trani (22.91%, 25.96%, and 4.35%, respectively), while Scenario 4 achieved the largest reduction in Wien (4.81%). Scenario 3 increased the ${\mathrm{CO}}_2$ emissions in all cases because gas boilers were employed to meet heating requirements.

In the following diagrams (Figure 8), the Annual Electric End Use is presented. More specifically, the diagrams show the percentage of electricity consumed for the operation of the fans and the heat pumps, the lighting, the space heating, and the cooling.

In Mallorca, for the present status of the building and Scenarios 1 and 2, the bar chart shows that the operation of fans consumes the most significant proportion of electricity, followed by lighting of the premises, and the consumptions of miscellaneous equipment, space heating, and cooling. For Scenarios 3 and 4, lighting of the spaces requires the most considerable proportion of electricity (32.9% and 29.3%, respectively), followed by exterior loads and miscellaneous equipment consumptions.

The bar graph for Palermo’s existing building condition and Scenarios 1 and 2 reveals that the operation of fans uses the greatest amount of power, followed by lighting the spaces and space heating and cooling. In Scenarios 3 and 4, the most substantial amount of electricity is required for space cooling and lighting. More specifically, in Scenario 3, 25.8% of electricity is used for space cooling and 25% for lighting, while in Scenario 4, 25.7% of electricity is consumed for lighting, followed by space cooling consumption (20.5%).

In Barletta-Andria-Trani, the diagram for the current building state and Scenarios 1 and 2 shows that space heating and lighting consume the most significant amount of electricity, followed by miscellaneous equipment consumption, fan operation, and space cooling. In Scenario 3, the most considerable percentage of electricity is used for the lighting of the premises (29%), followed by the miscellaneous equipment consumption (20.3%) and the space cooling consumption (19.7%). In Scenario 4, 30.3% of electricity is used for lighting, 21.3% for miscellaneous equipment, and 14.3% for the operation of pumps.

As far as Wien is concerned, the bar chart for the current building state and Scenario 1 shows that space heating consumes the largest percentage of electricity (25.6%), followed by the use of heat pumps (19.3%), lighting (18%), and fan operation (17%). In Scenario 3, lighting consumes 30.3% and miscellaneous equipment 21.3% of the electricity used. In Scenarios 2 and 4, the operation of heat pumps consumes the most considerable proportion of electricity, followed by lighting and miscellaneous equipment.

Figure 9 presents the percentage of fuel consumed for hot water and space heating. Regarding these charts, in the current state of the building and Scenarios 1, 2, and 4, 100% is consumed for domestic hot water production, as for the cover of cooling and heating loads, electricity is used. In Scenario 3, the percentages of fuel used for space heating and hot water production vary depending on the city. More specifically, in Mallorca, 66% of the fuel energy used is for room heating, and the remaining 34% goes toward producing hot water. In Palermo, 81.5% of the fuel is used for space heating, while 18.5% is used to provide hot water. In Barletta-Andria-Trani, 14.1% of the fuel is utilized for hot water production, and 85.9% is consumed for space heating. Finally, in Wien, 12.5% is consumed for the production of hot water and 87.5% for space heating. The difference in the percentages is due to the fact that every city studied has different heating needs to cover, with Wien and Barletta-Andria-Trani having the most significant heating needs. In contrast, Mallorca and Palermo have more minor heating needs.

The PROMETHEE II method was used to rank the scenarios in order to determine which of the recommended energy optimization strategies is optimal for each situation. The criteria chosen to evaluate the selected scenarios and then to derive their ranking are the following:

• Criterion 1: Annual Electricity Consumption (kWh/${\mathrm{m}}^2)$;
• Criterion 2: Annual Fuel Energy Consumption (kWh/${\mathrm{m}}^2)$;
• Criterion 3: Annual ${\mathrm{CO}}_2$ Emissions (Mg);
• Criterion 4: Energy Consumption’s Annual Cost (€/${\mathrm{m}}^2$).

After the PROMETHEE II method is applied, the scenarios are ranked, as shown in the following table (

Table 9. Scenario ranking table for Autodesk Revit, Insight, and Green Building Studio’s energy results.

	Ranking for Mallorca	Ranking for Palermo	Ranking for Barletta-Andria-Trani	Ranking for Wien
Scenario 1	2	2	3	3
Scenario 2	3	3	2	2
Scenario 3	4	4	4	4
Scenario 4	1	1	1	1

).

Based on the results of the multicriteria analysis methods, the best scenario for the cities studied is Scenario 4, in which the ASHRAE Terminal Package Heat Pump is used. Scenario 2 ranks comparatively high for Wien and Barletta-Andria-Trani, while scenario 1 is relatively high in the ranking for Mallorca and Palermo. As for scenario 3, it is fairly low in preference rankings for all the buildings studied.

4. Discussion

The building sector is responsible for the consumption of the largest percentage of energy produced and therefore plays a vital role in the depletion of natural resources, in the production of air pollutants, and generally in the pollution of the environment. Implementing saving measures in buildings can bring many economic and environmental benefits. Using software to simulate a building’s energy behavior is vital in order to identify the best procedures that result in energy savings.

This study performed an energy assessment of a university building in its existing state and provided four scenarios for energy optimization. After implementing the scenarios, the annual energy consumption, the annual energy costs for electricity and fuel, and the annual ${\mathrm{CO}}_2$ emissions were analyzed. For the modeling and energy evaluation of the current state of the University building, the ASHRAE Heat Pump was selected from the Autodesk Insight library. This heat pump has COP equal to 3.2 and EER equal to 9.5. For Scenario 1, 60% of the roof of the building was covered with photovoltaic panels (18.6% efficiency), and the thermal insulation of the roof was strengthened with 10 cm of thermal insulation material. In Scenario 2, a High-Efficiency Heat Pump was selected from the software’s library. This heat pump’s SEER is equal to 17.4, and the HSPF is equal to 9.6. Moreover, 60% of the building’s roof was covered with photovoltaic panels with an efficiency of 16%. For the modeling and energy evaluation of Scenario 3, a High-Efficiency VAV with a COP equal to 7.5 and a natural gas boiler with a combustion efficiency of 95% were selected. In addition, 75% of the building’s roof was covered with photovoltaic panels with an efficiency of 20.4%. In Scenario 4, the ASHRAE Terminal Package Heat Pump was selected with COP equal to 4.5 and EER equal to 11.9. Furthermore, 60% of the building’s roof was covered with photovoltaic panels, with an efficiency of 18.6%.

Subsequently, multicriteria analysis methods such as the AHP and PROMETHEE were used to determine the optimum intervention for each case. In more detail, the PROMETHEE II method rated the scenarios after the AHP method assigned the proper weights to the criteria examined based on their degree of importance. Along with PROMETHEE II ranking results, Scenario 4 was the best-case scenario in contrast to Scenario 3, which was the worst-case scenario.

Scenario 4, which was ranked as the best-case scenario for all cities studied, delivered a 35.65% reduction in yearly energy usage in Mallorca, a 40% reduction in yearly energy usage in Palermo, a 41.01% reduction in yearly energy usage in Barletta-Andria-Trani, and a 43.75% reduction in yearly energy usage in Wien. On this subject, it is worth noting that Kolokotsa et al. [9] obtained only a 30% decrease in energy use. In terms of yearly energy expenses, Scenario 4 reduced them from 30.05% to 47.31% in the cities studied. In terms of yearly power use, Scenario 4 reduced it from 35.58% to 43.36%, in contrast to Wai’s [39] research, which studied energy-saving measures for a Taiwanese university and only achieved a 16% reduction in annual energy usage. Scenario 4 lowered yearly fuel energy usage from 17.11% in Mallorca to 45.09% in Wien while also lowering annual ${\mathrm{CO}}_2$ emissions by up to 12.21% in Palermo.

Even though the interventions proposed can bring significant environmental, economic, and social benefits and increase the energy efficiency of the campus building, the Scenarios investigated could have been more extensive. The future goal of this research is the investigation of other sustainable HVAC systems, such as wind turbines, biomass boilers, and geothermal heat pumps, in order to explore the potential that these systems also offer to a university building in terms of energy optimization. It is also desirable to investigate buildings in additional cities to have a more comprehensive picture of potential energy savings.

5. Conclusions

The energy analysis results for educational buildings revealed that HVAC systems play the leading role in saving energy and reducing ${\mathrm{CO}}_2$ emissions for buildings of this size. For this reason, their initial choice in a newly erected building and their targeted replacement in a renovated building are essential. Equally important was integrating systems that exploit renewable energy sources (e.g., photovoltaic panels), which improve the energy class of the building and contribute to its sustainability.

The analysis of energy upgrading of a typical university building not only for one but four European cities with diverse climate conditions is one of the study’s key achievements. It is also worth mentioning that even though Building Information Modeling (BIM) tools have been used in the last few years to analyze buildings’ energy behavior, further research that gives an in-depth examination of university buildings’ energy needs was required. In contrast to previous research, in this study, some of the most popular BIM software programs (Autodesk Revit, Autodesk Insight, and Green Building Studio) were used in order to make a university energy analysis, presenting in detail all the features that BIM software programs can provide and how they can be used in decision-making in the energy sector. Another accomplishment of this study is that it established multicriteria approaches such as AHP and PROMETHEE as effective instruments for solving complicated and multidimensional issues such as choosing between multiple energy optimization scenarios. Throughout this study, all the steps to follow were analyzed, and all of the essential tools were provided for energy investigators to devise the optimum energy optimization intervention, considering the economic, technological, social, and environmental factors. This study’s case analysis suggests that associated energy researchers on university buildings can resort to these action plans in the future to adopt efficient energy-saving and carbon-reduction initiatives.