Energy Efficiency Optimization Model for Sustainable Campus Buildings and Transportation

Abstract

University campuses face significant challenges in balancing energy efficiency, renewable energy adoption, and sustainable transportation while meeting budgetary constraints and sustainability goals. While existing optimization approaches typically address these as separate problems, this study presents an innovative multi-objective optimization framework that integrates building efficiency, renewable energy, electric vehicle charging, and sustainability scoring criteria into a unified model. The approach formulates a mixed-integer non-linear programming model with three competing objectives: minimizing primary energy consumption, minimizing investment cost, and maximizing sustainability metrics, addressing the critical need for comprehensive campus energy management tools. The optimization model was applied to the R&D Park Building of Erciyes University, utilizing actual building parameters, time-variable electricity pricing, and commercially available renewable energy technologies. Our analysis of the Pareto-optimal solutions reveals distinct trade-offs between the objectives, with primary energy consumption ranging from 1,317,860 to 4,642,770 GJ/year, investment costs between $25,735 and $485,674, and sustainability scores between 366 and 1034. Most significant for practical implementation is the balanced performance solution ($127,064), which achieves minimum energy consumption (1,367,010 GJ/year) while securing a substantial sustainability score of 538 points. The results demonstrate that while inherent trade-offs exist between competing objectives, significant sustainability improvements are achievable at intermediate investment levels, making meaningful environmental progress accessible to a broad spectrum of higher education institutions. This comprehensive optimization framework provides campus administrators with a practical decision-support tool for aligning energy systems with institutional priorities, budgetary constraints, and sustainability commitments.

1. Introduction

The increase in primary energy consumption and climate change are among the most critical environmental problems of our time. Enhancing energy efficiency (EE) and integrating renewable energy sources (RES) are critical for combating climate change and ensuring energy security [1]. Growing energy demand and sustainable development goals necessitate the development of innovative solutions in the fields of EE and RES.

Higher education institutions represent significant energy consumers within the built environment. According to recent studies, Khoshbakht, Gou [2] documented typical campus consumption between 100 and 250 kWh/m² annually, while research-intensive universities can reach up to 400 kWh/m². Their analysis of 80 university buildings in Australia established that research facilities had the highest benchmark energy use intensity (216 kWh/m²/year), while academic offices showed the lowest (137 kWh/m²/year). This elevated consumption stems from diverse operational requirements and the finding that university buildings typically consume 2 to 3 times more energy than conventional office buildings due to extended hours, high-density occupancy, and specialized equipment needs. The environmental consequences of this consumption are substantial, with Robinson, Kemp [3] reporting that higher education institutions contribute approximately 1–2% of national carbon emissions in developed economies. Their study of 20 English universities found per-student emissions ranging from 1.27 to 4.9 tons of CO₂ annually, with 60–80% attributable to building energy consumption. Robinson, Kemp [3] identified concerning trends, including emissions increases in 18 of 20 research-intensive universities between 2005 and 2010, with some institutions like LSE and UCL showing alarming increases of 143% and 91%, respectively. Their work also revealed a paradox in carbon management: institutions setting realistic but conservative targets were often penalized in league tables and criticized for a lack of ambition, despite being more likely to achieve meaningful reductions. By 2010, the 20 English Russell Group institutions alone exceeded the entire sector’s 2020 emissions target, highlighting the magnitude of the challenge. With global higher education projected to grow from 250 million students in 2020 to 380 million by 2030 [4], developing comprehensive sustainability frameworks addressing both building energy systems and transportation has become both an environmental imperative and an operational priority for institutional management.

Various software tools have been developed for building energy efficiency optimization. For example, Diakaki, Grigoroudis [5] focused on reducing primary energy consumption, minimizing investment costs, and lowering CO₂ emissions with their multi-objective decision-making model. Later, Diakaki and Grigoroudis [6] presented an innovative approach that can flexibly incorporate decision-maker preferences into the model. Karmellos, Kiprakis [1] and Bayata and Temiz [7] developed MATLAB-based optimization tools. Penna, Prada [8] worked on optimal renovation solutions that promote nearly zero-energy building behavior. In recent research, Shi and Chen [9] presented an approach integrating automated machine learning and the NSGA-III algorithm. Vardopoulos, Santamouris [10] studied smart building technologies and Benaddi, Boukhattem [11] researched thermal management techniques. Abdou, El Mghouchi [12] demonstrated the potential of RES integration to increase EE in urban environments. These studies employ diverse methodological approaches, from multi-criteria decision analysis to simulation-based optimization, with varying degrees of experimental validation. However, most existing models address only specific applications rather than offering integrated solutions. This research gap necessitates a more versatile approach that incorporates environmental impacts and sustainability criteria alongside traditional energy efficiency and cost considerations. University campuses, with their diverse building types and substantial energy demands, provide ideal settings for developing and testing such comprehensive optimization frameworks, potentially yielding solutions applicable to broader community implementation.

Integrating RES and electric vehicles (EVs) into university campuses is crucial for sustainable development, serving both to reduce carbon footprints and create educational opportunities. Optimizing these systems—including PV installations, battery storage, and charging infrastructure—presents the design, energy management, and cost-effectiveness challenges that recent research has addressed through various approaches from system optimization to market integration strategies. Honarmand, Zakariazadeh [13] found a critical challenge posed by increasing EV adoption in power systems by developing a smart management and scheduling model for urban parking lots with large numbers of EVs. The study demonstrates that intelligent scheduling can achieve dual benefits, maintaining power system reliability while enabling EV owners to profit from vehicle-to-grid (V2G) services without compromising their desired state of charge at departure time. Building upon these optimization approaches for hybrid systems, Atia and Yamada [14] explored a different mathematical technique, a novel mixed-integer linear programming model for optimizing hybrid RES with battery storage in residential microgrids, with a particular focus on incorporating demand response from controllable appliances. The research demonstrates a sophisticated method for integrating modern power consumption patterns and eco-friendly technologies into microgrid design while maintaining computational efficiency. Shifting focus from general microgrid optimization to specific EV integration challenges, Farahmand, Nazari [15] focused on optimizing the design of hybrid PV-battery-diesel systems that combine solar (PV), wind, and diesel generation with battery storage for off-grid residential and small business applications. The work demonstrates how standalone hybrid systems can become both economically viable and environmentally beneficial, particularly as renewable technology costs decrease. While Farahmand, Nazari [15] employed Genetic Algorithm optimization, Ramli, Bouchekara [16] advanced the optimization techniques by examining hybrid microgrid systems that combine solar, wind, and diesel generators with battery storage.

The literature shows a progression in EV parking lot management research. Modarresi Ghazvini and Olamaei [17] demonstrated that integrating V2G parking lots into hybrid energy systems can reduce total system costs by 5.21% through an optimization algorithm. Later, Osório, Lotfi [18] advanced this concept by proposing solar-powered EV parking lots that can participate profitably in both energy and ancillary services markets. This evolution reflects a shift from viewing EV parking lots merely as cost-reduction tools to recognizing them as active market participants capable of providing essential grid services. Pai, Senjyu [19] demonstrated an innovative home energy management system integrating PV, wind turbines, and hybrid backup storage (including hydrogen storage, batteries, and EVs with V2H technology), showing that implementing demand response programs significantly reduced peak load and operating costs, with the 40% DR scenario achieving a 2.34% reduction in operating costs compared to no DR implementation. Fachrizal, Shepero [20] synthesized current research on smart charging strategies that integrate PV systems with electric vehicles while considering existing electricity consumption patterns to provide a comprehensive overview of these developments. The work demonstrates multiple aspects of smart charging, including various objectives, such as improving PV utilization and reducing peak loads, control configurations (centralized versus distributed), and implementation settings (homes and workplaces). They analyze both optimization techniques and rule-based algorithms used in smart charging schemes. They conclude that future research should focus on finding practical balance points between system complexity and performance, particularly regarding control configurations, charging algorithms, and forecast integration.

Universities are pivotal in shaping sustainable development through education, research, and campus operations. The UI GreenMetric World University Rankings, established in 2010, has emerged as a significant tool for evaluating and comparing university sustainability efforts globally [21]. This ranking system assesses various aspects of campus sustainability, including setting and infrastructure, energy and climate change, waste management, water resources, transportation, and education. As higher education institutions increasingly recognize their responsibility in addressing environmental challenges, sustainable campus initiatives have become crucial for reducing environmental impact, cutting operational costs, and providing practical learning opportunities for students. These initiatives contribute to global sustainability goals and enhance institutional reputation and competitiveness. Recent studies have demonstrated various approaches to implementing and measuring campus sustainability, from comprehensive frameworks to specific technological solutions such as renewable energy integration and smart energy management systems. The growing emphasis on campus sustainability reflects a broader understanding that universities must lead by example in the transition toward a more sustainable future. Heravi, Aryanpour [22] developed and implemented a comprehensive green university framework at the University of Tehran, using statistical techniques to identify and prioritize key practice areas. Through structural equation modeling, they identified water, energy, sustainable behavior, and procurement as critical areas, with sustainable behavior and procurement being the most fundamental based on influence-dependency analysis. The implementation of this framework led to significant improvements in the university’s UI GreenMetric scores, particularly in Setting and Infrastructure (218%) and Education (165%), with an overall improvement of 64% between 2016 and 2020, demonstrating the effectiveness of tailoring green practices to university-specific characteristics. Building upon frameworks for sustainability measurement, Jiang and Kurnitski [23] developed the PICSOU framework with approximately 20 core metrics that track campus emissions and environmental impacts. Their application revealed optimization opportunities in building space utilization and parking area reduction, measuring campus carbon footprints at approximately 1.3 tons per person annually. While earlier studies focused on frameworks and measurement tools, Diniz, da Silva [24] investigated specific technological implementations, studying PV systems at the PUC Minas University campus in Brazil, where renewable energy comprises 82% of the national energy mix. The researchers developed a student-led methodology for campus solar integration that included power consumption assessment, solar resource evaluation, and optimization of panel placement according to Brazilian standards. The study demonstrated that PV implementations simultaneously reduced emissions, decreased costs, provided learning opportunities, and improved institutional rankings through alignment with UN Sustainable Development Goals. This evolution from conceptual frameworks to practical technological implementations highlights the value of standardized assessment tools like UI GreenMetric, enabling universities to pursue sustainability goals while maintaining academic excellence.

1.1. Multi-Objective Building Optimization

This study builds upon the framework established by Karmellos, Kiprakis [1], developing a multi-objective approach for the optimal prioritization of energy efficiency measures in buildings. Their model addresses the fundamental challenge of balancing primary energy consumption reduction with investment cost minimization, formulated as a mixed-integer non-linear programming (MINLP) problem. The model can be characterized by its two main objective functions: (1) minimization of total primary energy consumption and (2) minimization of total investment cost:

$$\min \left[g_1\left(x\right)\right]=Q_T$$

(1)

$$\min \left[g_2\left(x\right)\right]=INVCOST$$

(2)

where $Q_T$ represents the total annual primary energy consumption (GJ/year) and $INVCOST$ represents the investment cost (€).

1.2. Expanded Framework for Campus Applications

While the model by Karmellos, Kiprakis [1] provides a foundation for building-level optimization, university campuses present additional complexities that require an expanded framework. Campus settings typically incorporate spatial constraints for renewable energy installations, complex energy demand patterns, and institutional sustainability goals that extend beyond individual building performance.

Building upon these foundational concepts, this research extends the Karmellos, Kiprakis [1] framework to address the unique sustainability challenges of university campuses. While the original model excels at building-level optimization, the interconnected nature of campus environments requires a more comprehensive approach that integrates additional dimensions beyond energy consumption and cost.

To address the specific requirements of university campus environments, we have expanded the Karmellos, Kiprakis [1] model in three critical dimensions as follows:

• Solar Energy Production and Management: This research incorporates the optimization of PV system installation and operation, accounting for campus-specific constraints such as roof availability, orientation, and grid interaction. A key feature is the implementation of monthly settlement calculations that account for variable feed-in tariffs and time-of-use electricity pricing, enabling accurate economic assessment of grid interactions. This addition allows the model to balance energy efficiency improvements with on-site renewable energy generation.
• Electric Vehicle Charging Infrastructure: Recognizing the growing importance of sustainable transportation in campus settings, our model includes the optimization of EV charging stations. This component addresses station placement, capacity planning, and charging schedule to minimize grid impact while maximizing service availability.
• UI GreenMetric Scoring Integration: The UI GreenMetric World University Ranking system has emerged as a global standard for evaluating campus sustainability. We have integrated this framework as an additional objective function, allowing institutions to explicitly optimize toward improved sustainability rankings while balancing energy and cost considerations. The third objective function, sustainability score maximization, is as follows:

  $$max\left[g₃\left(x\right)\right]=UIGMS$$

  (3)

  where UIGMS represents the sustainability score. The additional objective is integrated into the multi-objective optimization framework, allowing decision-makers to balance sustainability metrics with energy efficiency and economic considerations according to institutional priorities.

These enhancements transform the building-centric approach of the original model into a comprehensive campus energy optimization framework that addresses both stationary energy use in buildings and the dynamic energy requirements of transportation systems. By incorporating these improvements, our model better aligns with the unique needs of university campus buildings, offering a more precise and effective optimization solution.

The mathematical formulation of the campus-specific model builds upon the equations developed by Karmellos, Kiprakis [1], with additional formulations to address the expanded scope. The complete set of equations for building energy optimization is provided in Appendix A, while the campus-specific extensions are detailed in the following sections.

Integrating RES and EV into university campuses is essential for creating sustainable and smart campus environments. Higher education institutions increasingly implement these technologies to reduce carbon footprints, demonstrate environmental leadership, and create living laboratories for research and education. These integrated systems present both challenges and opportunities in system design, energy management, and cost-effectiveness that require sophisticated optimization approaches. Previous optimization approaches in this domain exhibit several critical limitations. First, most existing models, such as those developed by Diakaki, Grigoroudis [5] and Bayata and Temiz [7], focus exclusively on building-level interventions without addressing the interconnected nature of campus energy systems. As Robinson, Kemp [3] demonstrated, the fundamental issue with many carbon management plans is their reliance on unrealistic target-setting rather than achievable, pragmatic goals. Their analysis revealed that 90% of research-intensive universities in England were consistently increasing emissions while simultaneously setting ambitious reduction targets. The disconnect between targets and reality highlights a significant flaw in current approaches. Second, transportation electrification is typically treated as a separate optimization problem from building energy management, as evident in the works by Honarmand, Zakariazadeh [13] and Modarresi Ghazvini and Olamaei [17], preventing the identification of synergistic solutions that leverage shared infrastructure. Third, existing frameworks by Heravi, Aryanpour [22] and Jiang and Kurnitski [23] lack the quantitative integration of internationally recognized sustainability metrics into the optimization process, relying instead on post hoc evaluation of proposed solutions against sustainability criteria. Finally, most models employ single-objective optimization approaches or weighted-sum multi-objective methods that fail to characterize the inherent trade-offs between competing objectives adequately. This study addresses these limitations by developing a comprehensive MINLP framework that integrates these previously siloed domains into a unified optimization approach.

Recent research has demonstrated that optimizing these systems requires addressing both technical and economic considerations through innovative mathematical models and strategies that balance multiple competing objectives.

2. Proposed Method

The optimization model developed for sustainable campus energy systems integrates three critical components into the building efficiency model: solar energy production, electric vehicle charging infrastructure, and UI GreenMetric scoring criteria. This integration creates a comprehensive approach that addresses both the stationary energy needs of campus buildings and the dynamic requirements of transportation systems. The nomenclature for the model is presented in

Table 1. Nomenclature.

Symbol	Description	Type	Unit
sp	Solar panel type (550 W, 450 W, 375 W)	Set	-
area	Areas for solar panel installation (roof, parking)	Set	-
ho	Hours of the day	Set	-
day_hours(ho)	Day hours	Set	-
night_hours(ho)	Night Hours	Set	-
inv	Inverter types (30 kW, 20 kW, 15 kW, 8 kW)	Set	-
m	Months	Set	-
$APANE{L}_{sp}$	Area of one solar panel	Input Parameter	m2
$A_{area}$	Maximum available roof and park area	Input Parameter	m2
$BLCRef$	Reference value for building load coefficient	Input Parameter	W/K
$CHD$	Time needed to fully charge an EV	Input Parameter	hour
$CHEF$	EV charging efficiency	Input Parameter	%
$CHPD$	Available hours for charging per day	Input Parameter	hour
$CPL$	Maximum charging power per station	Input Parameter	W
$CRSCST$	Cost per charging station installation	Input Parameter	$
$CST{P}_{sp}$	Cost per solar panel	Input Parameter	$
$EAP$	UI GreenMetric points for Energy efficient appliances	Input Parameter	-
$ECP$	Energy conservation points of UI GreenMetric	Input Parameter	-
$EERP$	Renewable energy ratio points of UI GreenMetric	Input Parameter	-
$ERP$	Renewable energy points of UI GreenMetric	Input Parameter	-
$EVBC$	Average EV battery capacity	Input Parameter	Wh
$EVP$	The total number of Zero Emission Vehicles (ZEV) divided by total campus population points	Input Parameter	-
$EVPP$	The total number of vehicles divided by the total campus population points of UI GreenMetric	Input Parameter	-
$EVZP$	Zero emission vehicles points of UI GreenMetric	Input Parameter	-
$FEE{D}_m$	Feed-in tariff for excess solar power	Input Parameter	$
$HE{P}_{ho,m}$	Electricity price for each hour and month	Input Parameter	$
$HS{R}_{ho,m}$	Solar radiation for each hour and month	Input Parameter	$
$ICST$	Installation cost per m2	Input Parameter	$
$INE{F}_{inv}$	Efficiency of each inverter type	Input Parameter	%
$INC{P}_{inv}$	Capacity of each inverter type	Input Parameter	W
$INCS{T}_{inv}$	Cost of each inverter type	Input Parameter	$
$MACST$	Monitoring system costs per year	Input Parameter	$
$MAXPARK$	Maximum number of vehicles in the parking lot	Input Parameter	-
$MAXPANEL$	Maximum possible number of panels that can be installed	Input Parameter	-
$MCST$	Annual maintenance cost per m2	Input Parameter	$
$PE{F}_{sp}$	Efficiency of solar panels	Input Parameter	%
$SL$	System losses (Inverter losses, wiring losses, soiling/dirt losses, shading losses, temperature losses)	Input Parameter	-
$ASG$	Total annual solar generation	Calculated Parameter	Wh
$CR{S}_{area}$	Maximum number of charging stations that can be installed in each area	Calculated Parameter	-
$EAS$	Energy efficient appliances score	Calculated Parameter	-
$ECS$	Energy conservation score	Calculated Parameter	-
$ERRS$	Score for renewable energy ratio	Calculated Parameter	-
$ERS$	Renewable energy score	Calculated Parameter	-
$EVPS$	The total number of vehicles (cars and motorcycles) divided by the total campus’ population	Calculated Parameter	-
$EVC{H}_{ho,m}$	Number of EVs charging in each hour	Calculated Parameter	-
$EVZS$	Zero-emission vehicles score	Calculated Parameter	-
$Q_{EV}$	Energy consumption of electric vehicles	Calculated Parameter	-
$HPA{P}_{ho,m}$	Power used for appliances	Calculated Parameter	Wh
$HP{G}_{ho,m}$	Solar power generated in each hour of each month	Calculated Parameter	Wh
$HPL{I}_{ho,m}$	Power used for lighting	Calculated Parameter	Wh
$HP{U}_{ho,m}$	Electricity used for EV charging in each hour	Calculated Parameter	Wh
$MGCS{T}_m$	Net grid cost/profit after settlement for each month	Calculated Parameter	$
$TEU$	Total annual energy usage	Calculated Parameter	Wh
$UIGMS$	Total sustainability score	Calculated Parameter	-
$DEVT$	Number of EVs to charge per day	Decision Variables	-
$INVI{N}_{inv}$	Number of each type of inverter installed	Decision Variables	-
$PI{T}_{area,sp}$	Number of each type of solar panel installed in each area	Decision Variables	-
$HP{B}_{ho,m}$	Power bought from the grid	Decision Variables	Wh
$HP{S}_{ho,m}$	Power sold to the grid	Decision Variables	Wh

.

2.1. Renewable Energy System Cost and Constraints

The total cost associated with RES is calculated using Equation (4), which provides a holistic accounting of all relevant expenses:

$$\begin{gathered} COS{T}_{RES}={\displaystyle \sum }_{area=1}^{AREA}{\displaystyle \sum }_{sp=1}^{SP}\left(PI{T}_{area,sp}CST{P}_{sp}+APANE{L}_{sp}ICST+PI{T}_{area,sp}APANE{L}_{sp}MCST\right)+ \\ {\displaystyle \sum }_{inv=1}^{INV}\left(INVI{N}_{inv}INCS{T}_{inv}\right)+MACST+{\displaystyle \sum }_{area=1}^{AREA}\left(CR{S}_{area}CRSCST\right)+{\displaystyle \sum }_{m=1}^M\left(MGCS{T}_m\right) \end{gathered}$$

(4)

The equation captures the complete economic picture by incorporating direct equipment costs (solar panels and inverters), infrastructure costs (installation and charging stations), ongoing expenses (maintenance and monitoring), and net operational costs from grid interactions. Our approach differs from previous models by considering the spatial distribution of renewable resources across multiple campus areas and their interaction with transportation infrastructure. The physical implementation constraints are governed by Equations (5) and (6):

$${\displaystyle \sum }_{sp=1}^{SP}\left(PI{T}_{area,sp}APANE{L}_{sp}\right)\le A_{area}$$

(5)

$${\displaystyle \sum }_{inv=1}^{INV}\left(INVI{N}_{inv}INC{P}_{inv}INE{F}_{inv}\right)\ge HP{G}_{ho,m} \forall ho \in H, \forall m \in M$$

(6)

These constraints realistically model the campus environment by accounting for the limited available space for renewable installations and ensuring proper sizing of power conversion infrastructure.

2.2. Energy Balance and Generation

The model incorporates a sophisticated approach to energy balance that accounts for the temporal dynamics of solar generation, building loads, and EV charging patterns. The hourly solar power generation is calculated using Equation (7):

$$HP{G}_{ho,m}={\displaystyle \sum }_{area=1}^{AREA}{\displaystyle \sum }_{sp=1}^{SP}\left(PI{T}_{area,sp}APANE{L}_{sp}HS{R}_{ho,m}PE{F}_{sp}SL\right)$$

(7)

This equation calculates the hourly solar power generation by summing the contribution of all panel types across all installation areas, accounting for panel area, solar radiation, panel efficiency, and system losses. This generation is integrated into the energy balance framework through Equation (8):

$$HP{G}_{ho,m}+HP{B}_{ho,m}-HP{U}_{ho,m}-HPA{P}_{ho,m}-HPL{I}_{ho,m}-HP{S}_{ho,m}=0$$

(8)

This energy balance equation ensures that at any given hour, the sum of power generated and purchased equals the sum of power consumed (by EVs, appliances, and lighting) and sold to the grid. A notable aspect of this approach is the integration of building energy systems with both renewable generation and EV charging demand. This integration creates the potential for a coordinated operation that enhances the model’s utility compared to building-only frameworks. The economic impact of this energy flow is quantified through the monthly settlement calculation in Equation (9):

$${\displaystyle \sum }_{m=1}^M\left(MGCS{T}_m\right)={\displaystyle \sum }_{ho=1}^{HO}HP{B}_{ho,m}HE{P}_{ho,m}-{\displaystyle \sum }_{ho=1}^{HO}HP{S}_{ho,m}FEE{D}_m \forall m \in M$$

(9)

This equation calculates the monthly net cost or profit from grid interactions by subtracting revenue from electricity sales from the cost of electricity purchases, using time-variable pricing. To properly account for the building energy demand patterns, we distribute consumption for appliances and lighting across appropriate hours using Equations (10) and (11):

$$HPA{P}_{ho,m}=Q_m^{AD}{\displaystyle \frac{1000}{FCONV card\left(DAYH\right)}} \forall ho \in DAYH \forall m \in M$$

(10)

$$HPL{I}_{ho,m}=Q_m^{LD}{\displaystyle \frac{1000}{FCONV card\left(NIGHTH\right)}} \forall ho \in NIGHTH \forall m \in M$$

(11)

These equations distribute the appliance and lighting loads across the appropriate day and night hours, respectively, ensuring an accurate representation of daily consumption patterns.

The PV system optimization considers a comprehensive electricity demand profile for the campus building, including lighting systems and the energy required for computers, laptops, and displays. As Khoshbakht, Gou [2] identified in their study of university buildings, electricity consumption varies significantly by activity type and discipline, with research spaces consuming up to three times more energy than office spaces. Our model employs an offset method to balance energy production and consumption, where the solar panels are installed according to our multi-objective optimization function based on the building’s energy requirements. The energy produced from these panels is fed directly into the grid, with monthly settlement calculations performed to determine net energy costs or credits. This grid-interactive approach enables more flexible system sizing since energy production and consumption do not need to be temporally matched. While our current implementation focuses on lighting and computing equipment loads, the model framework is designed to accommodate additional energy consumers if necessary, providing scalability from individual building applications to campus-wide implementations with multiple load centers.

2.3. Electric Vehicle Charging Infrastructure

A significant advancement in our model is the comprehensive treatment of campus EV charging infrastructure. This component addresses the growing importance of sustainable transportation in university settings through a series of interconnected constraints and calculations. The operational boundaries of the charging system are established by Equations (12) and (13), which ensure that the system accounts for both physical infrastructure limitations (available charging stations) and electrical capacity constraints:

$$EVC{H}_{ho,m}\le {\displaystyle \sum }_{area=1}^{AREA}\left(CR{S}_{area}\right)$$

(12)

$$HP{U}_{ho,m}\le EVC{H}_{ho,m}CPL$$

(13)

The model also incorporates practical implementation considerations for EV charging infrastructure. Equations (14) and (15) address specific real-world deployment factors:

$$CR{S}_{’roo{f}^{\prime }}=0$$

(14)

$${\displaystyle \frac{CR{S}_{’parkin{g}^{\prime }}CHPD}{CHD}} \ge DEVT$$

(15)

Equation (14) reflects safety and accessibility requirements by prohibiting rooftop charging stations, while Equation (15) ensures sufficient charging capacity to meet the daily EV demand of the campus community. The daily energy requirement calculation in Equation (16) connects the transportation and energy systems:

$${\displaystyle \sum }_{ho=EVH}^{HO}HP{U}_{ho,m}=DEVT{\displaystyle \frac{EVBC}{CHEF}} \forall ho \in DAYH \forall m \in M$$

(16)

Equation (16) establishes the connection between EV charging requirements and the campus energy balance, enabling a comprehensive analysis of the interactions between transportation electrification and energy management.

2.4. UI GreenMetric Scoring Integration

Another innovative aspect of our model is the incorporation of the UI GreenMetric framework as an explicit optimization objective. This integration allows universities to align their energy and transportation planning directly with internationally recognized sustainability metrics. The total sustainability score is calculated using Equation (17):

$$UIGMS=ERRS+ERS+EAS+ECS+EVPS+EVZS$$

(17)

The unified scoring approach encompasses multiple dimensions of campus sustainability, creating a comprehensive framework for evaluation and optimization. The model calculates individual components using specific criteria that reflect both the technical performance of systems and their contribution to campus sustainability. The following equations evaluate both the relative contribution of renewables to the campus energy mix and the utilization of available space for renewable infrastructure:

$$ERRS=EERP {\displaystyle \frac{ASG}{TEU}}$$

(18)

$$ERS=ERP{\displaystyle \frac{{{\displaystyle \sum }}_{area=1}^{AREA}{{\displaystyle \sum }}_{sp=1}^{SP}\left(PI{T}_{area,sp}\right)}{MAXPANEL}}$$

(19)

Our approach extends beyond renewable generation to include energy efficiency measures, rewarding improvements in both equipment selection and building envelope performance as follows:

$$EAS=EAP\left(1-{\displaystyle \frac{{{\displaystyle \sum }}_{eaj=1}^{EAJ}(x_{eaj}^{EA}{f}_{load,eaj})}{f_{loadMAX,eaj}}}\right)$$

(20)

$$ECS=ECP {\displaystyle \frac{BLCRef}{BLC}}$$

(21)

The transportation metrics in the model create incentives for reducing overall vehicle dependency on campus while increasing the proportion of zero-emission vehicles with the following equations:

$$EVPS=EVPP \left(1-{\displaystyle \frac{MAXPARK}{n_{people}}}\right)$$

(22)

$$EVZS=EVZP {\displaystyle \frac{DEVT}{MAXPARK}}$$

(23)

The model ties sustainability metrics to concrete energy performance through Equations (24) and (25):

$$ASG={\displaystyle \sum }_{ho=1}^{HO}{\displaystyle \sum }_{m=1}^M\left(HP{G}_{ho,m}{\displaystyle \frac{t{d}_m}{1000}}\right)$$

(24)

$$TEU=\left(Q_L+Q_A\right)0.277778+{\displaystyle \sum }_{ho=1}^{HO}{\displaystyle \sum }_{m=1}^M\left(HP{U}_{ho,m}{\displaystyle \frac{t{d}_m}{1000}}\right)$$

(25)

These calculations provide the quantitative foundation for sustainability metrics, ensuring that the model’s recommendations are based on actual energy performance rather than abstract scoring criteria.

Equation (26) introduces a critical addition to the model by explicitly quantifying the total energy consumption of electric vehicles ($Q_{EV}$). By incorporating this parameter, the model provides a more comprehensive accounting of the campus energy ecosystem, capturing not only traditional building energy use but also the additional load introduced by electrified transportation.

$$Q_{EV}={\displaystyle \sum }_{ho=1}^{HO}{\displaystyle \sum }_{m=1}^M\left(HP{U}_{ho,m}{\displaystyle \frac{t{d}_m}{1000}}\right)FCONV{\displaystyle \frac{f^{grid}}{n_{grid}}}$$

(26)

The mathematical framework represents a significant advancement in campus energy optimization by integrating building performance, renewable generation, electric transportation, and sustainability metrics into a cohesive system. This approach enables campus administrators to identify truly optimal configurations that balance technical, economic, and environmental priorities according to the institution’s specific needs and constraints. Detailed equations for building energy calculations are provided in Appendix A.

2.5. Model Limitations and Assumptions

The optimization framework presented in this study, while extensive in scope, operates under several assumptions that represent potential limitations. In the Solar Energy Production and Management component, the model assumes consistent panel performance degradation across all installation areas and does not account for microclimate variations that might affect yield differently on rooftop versus parking installations. The temporal resolution of solar radiation data (hourly averages) may smooth out short-duration fluctuations that could impact actual system performance.

The model also employs a simplified approach to grid interaction that assumes unlimited grid capacity for both the import and export of electricity, potentially overlooking constraints that might exist in actual implementation scenarios. Additionally, while the model accounts for basic system losses, it does not incorporate detailed modeling of partial shading impacts or soiling patterns that vary seasonally.

From an economic perspective, the framework employs static cost parameters throughout the optimization process, whereas actual market conditions for renewable energy components demonstrate significant volatility. This simplification may impact the robustness of economic projections, particularly for multi-year implementation scenarios.

These limitations suggest directions for future research, including the incorporation of stochastic optimization approaches to address uncertainties in solar resource availability, more detailed modeling of microclimatic conditions, and dynamic pricing models for both technology costs and grid interactions.

3. Case Study

A case study for the model is performed for the R&D Park Building of Erciyes University. This sample building comprises three blocks and is actively utilized by academics and students. The building has a total volume of 13,047.44 m³, with 1279.407 m² of wall area (excluding windows and doors), 1412.41 m² of floor area, and an equal amount of ceiling area. The building features 350 windows (each 0.7 m²) and 1 entrance door (6.89 m²). The average occupancy is 110 people. For computational efficiency, the optimization calculation is performed for a single input block, illustrated in Figure 1.

The building envelope components include external walls with U-values of brick of 0.45 W/m²K, coat of 0.51 W/m²K, and plaster of 0.7 W/m²K, standard double-glazed windows with a U-value of 2.7 W/m²K before optimization, and a flat roof with a U-value of 0.72 W/m²K. The existing door has a U-value of 3.1 W/m²K. The optimization model considers improvements to these envelope components through various options, including enhanced insulation materials, high-performance glazing, and improved door systems, with potential U-values ranging from 0.2–1.6 W/m²K for walls, 1.6–5 W/m²K for windows, and 2.1–4.0 W/m²K for doors.

3.1. Parameters

The optimization model parameters are categorized into five main groups to address the comprehensive campus energy ecosystem. First, Energy Demand Calculation Parameters include environmental factors, building envelope properties, and occupancy patterns. Second, Primary Energy Consumption Parameters encompass electrical equipment specifications and usage patterns. Third, Renewable Energy System Parameters include solar panel specifications (550 W, 450 W, and 375 W), panel efficiency ratings, and available installation areas (roof and parking). Fourth, Electric Vehicle Infrastructure Parameters comprise charging station capacities, efficiency factors, average battery capacity, and daily charging patterns. Fifth, UI GreenMetric Parameters incorporate point allocations for energy conservation, the renewable energy ratio, energy-efficient appliances, and zero-emission vehicles. This comprehensive parameterization enables the holistic optimization of the campus energy system while balancing sustainability metrics and economic considerations.

Table 2. Key characteristics of solar panels.

Type	Area (m2)	Efficiency (%)	Cost (USD)
Panel-1 (550 W)	2.58	0.216	136
Panel-2 (450 W)	2.18	0.206	112
Panel-3 (375 W)	1.82	0.205	95

and

Table 3. Key characteristics of inverters.

Type	Capacity (W)	Efficiency (%)	Cost (USD)
Inverter-1	30,000	1	7289
Inverter-2	20,000	1	5223
Inverter-3	15,000	1	4419
Inverter-4	8000	1	3415

present the technical specifications and economic parameters for the solar panels and inverters considered in this study. Table 2 summarizes the key characteristics of three solar panel options (550 W, 450 W, and 375 W), including their physical dimensions, efficiency ratings, and associated costs. Table 3 details the specifications of four inverter types (30 kW, 20 kW, 15 kW, and 8 kW), highlighting their capacity, efficiency, and investment requirements. While the building-related parameters and energy efficiency measures are adopted from Movlyanov, Akyol [25], the renewable energy system and electric vehicle infrastructure parameters presented in these tables are newly developed for this study. This combination of established building parameters with newly created renewable energy and EV data allows the model to represent the integration challenges specific to campus environments accurately. The parameter sets reflect commercially available technology options, ensuring the optimization results provide realistic implementation guidance. By including multiple component options with varying capacities and cost profiles, the model can identify optimal technology combinations for different institutional priorities and constraints.

3.2. Multi-Objective Optimization Approach

We employ a multi-objective optimization approach to balance the competing objectives in campus energy system design. This methodology addresses the complex challenge of simultaneously optimizing multiple criteria, including energy efficiency, renewable energy integration, and UI GreenMetric scoring. Unlike single-objective approaches that combine multiple criteria into a composite function, this approach preserves the distinct nature of each objective, allowing decision-makers to evaluate trade-offs directly [26].

The model is formulated as a mixed-integer non-linear programming (MINLP) problem using the General Algebraic Modeling System (GAMS), which serves as the standard interface for the Neos server solvers. BARON and LINDO solvers are utilized to achieve local optimal solutions at each sampling time, from which Pareto curves are extracted to visualize the relationships between energy parameters, cost considerations, and sustainability metrics. This visualization of trade-offs provides campus planners with comprehensive insights into decision-making.

4. Results

The multi-objective optimization model for campus energy systems was applied to the R&D Park Building of Erciyes University. The model considers the integration of PV systems, EV charging infrastructure, and building EE measures while optimizing the investment cost, primary energy consumption, and UI GreenMetric sustainability score.

The optimization process yielded a Pareto set of solutions that represents the trade-offs between three competing objectives. Figure 2 presents the Pareto solutions, demonstrating the inherent trade-offs between these objectives. Each point on this surface represents a unique configuration of the campus energy system.

The analysis reveals that primary energy consumption and investment cost exhibit a competing relationship. As shown in Figure 2, solutions with lower primary energy consumption generally require higher initial investments. However, the integration of the UI GreenMetric scoring introduces an additional dimension to this relationship, creating regions of the solution space where moderate investments can achieve significant sustainability scoring improvements. As presented in

Table 4. Minimum and maximum values of objective functions.

Solution	Primary Energy Consumption (GJ/year)	Investment Cost (USD)	Sustainability Score
Minimum energy solution	1,317,860	174,963	366.2
Minimum cost solution	4,642,770	25,735	452.2
Maximum sustainability solution	1,764,160	485,674	1033.8
A balanced solution	1,367,010	127,064	538.3

, the primary energy consumption ranges from 1,317,860 to 4,642,770 GJ/year, while investment costs vary from $25,735 to $485,674, and sustainability scores span from 366.2 to 1033.8 points, illustrating the breadth of possible system configurations available to decision-makers.

Decision makers can utilize the Pareto surface presented in Figure 2 as a practical decision-support tool by first identifying acceptable thresholds for each objective based on institutional constraints and priorities. For example, budget-constrained institutions might set a maximum investment threshold of $150,000, and then examine the achievable combinations of energy consumption and sustainability scores within that constraint. The clustering of solutions in certain regions of the surface reveals ‘sweet spots’ where marginal improvements in one objective can be achieved with minimal sacrifice to others. The moderate investment range ($100,000–$150,000) is particularly noteworthy, as it contains solutions that balance all three objectives effectively, making them attractive starting points for phased implementation. Decision-makers can also identify diminishing returns—points beyond which additional investment yields minimal improvement in either energy consumption or sustainability score.

Figure 3 illustrates the relationship between investment cost and primary energy consumption across different optimization solutions. The distinct pattern reveals that higher investment costs generally correspond to lower energy consumption, demonstrating the trade-off between financial commitment and energy efficiency. The solutions clustered at higher investment levels achieve the lowest energy consumption values, indicating investments in energy-efficient building envelope components, high-performance heating systems, and efficient hot water systems. Conversely, the solutions with minimal investment result in significantly higher energy consumption. The moderate investment range ($100,000–$150,000) shows a cluster of solutions that provide substantial energy savings compared to the baseline, representing potentially cost-effective compromise solutions. This visualization clearly demonstrates that while significant energy efficiency improvements require financial investment, there are diminishing returns at the highest investment levels.

Figure 4 depicts the relationship between energy consumption and sustainability score across the solution space. The analysis indicates that achieving the maximum sustainability score of 1034 points necessitates a substantial rise in investment costs without an improvement in energy efficiency. This correlation stems from the requirement to convert all 50 vehicles in the campus parking facility to electric vehicles and establish the corresponding charging infrastructure. The optimal configuration for maximizing the sustainability score involves comprehensive solar PV coverage of both the building rooftop and parking area, creating an integrated RES that supports both building operations and transportation electrification.

Figure 5 illustrates the trade-off between investment cost and sustainability score. The optimization results demonstrate that maintaining minimal investment costs while achieving elevated sustainability scores leads to increased energy consumption. This trade-off emerges because the objectives cannot be simultaneously optimized to their individual extremums. The Pareto-optimal solutions reveal that balancing cost constraints with sustainability metrics requires compromises in energy efficiency, particularly when investment resources are limited, but sustainability scoring remains a priority.

The capital expenditure for the optimized solutions ranges from $120,000 to $350,000, with the solar PV system typically representing the largest component (45–60% of total cost), followed by EV charging infrastructure (20–35%) and EE upgrades (15–25%). The solutions at the lower end of the investment spectrum focus primarily on low-cost EE measures and limited PV deployment. In contrast, higher-investment solutions incorporate comprehensive RES and extensive EV charging infrastructure.

A sensitivity analysis was conducted to evaluate the robustness of the optimization results to variations in key parameters, including electricity prices, solar panel costs, and EV adoption rates. The results demonstrate that the model produces stable solutions across reasonable variations in these parameters, with solar PV deployment remaining economically advantageous even under conservative cost projections. However, the optimal size of EV charging infrastructure is more sensitive to assumptions about EV adoption rates, indicating that a phased implementation approach may be prudent for this component.

The sensitivity analysis revealed that electricity prices had the most significant impact on the optimal solution, particularly influencing the economic viability of renewable energy installations. When electricity prices increased, the model showed a stronger preference for solar panel deployment, in contrast to conventional expectations. This is because the government supports mechanisms for renewable energy to create favorable economic conditions when grid electricity becomes more expensive, effectively incentivizing greater renewable capacity. Solar panel costs demonstrated moderate sensitivity, with decreased costs predictably leading to increased installation capacity, though the relationship was not strictly linear due to space and system integration constraints. Interestingly, EV adoption rates showed the least direct impact on energy optimization parameters, primarily affecting the sustainability score rather than energy consumption or cost metrics. Increases in EV adoption were primarily driven by sustainability goals rather than energy or economic optimization, highlighting the importance of the multi-objective approach that includes UI GreenMetric scoring as a distinct objective function. This finding substantiates the conclusion that transportation electrification decisions in campus environments are more strongly influenced by institutional sustainability commitments than by pure energy economics.

To illustrate key trade-offs identified by the multi-objective optimization, representative solutions from the Pareto front are presented in

Table 5. Representative solutions.

Primary Energy Consumption (GJ/year)	Investment Cost (USD)	Sustainability Score (points)	Primary Energy Consumption (GJ/year)	Investment Cost (USD)	Sustainability Score (points)
1,317,860	146,271	366	1,367,010	127,064	538
1,317,860	173,620	551	1,368,210	119,604	383
1,317,860	187,049	591	1,764,160	485,674	1034
1,317,860	210,746	616	1,958,360	111,849	511
1,317,860	223,806	630	2,131,070	104,684	500
1,317,860	241,914	649	2,173,300	94,784	289
1,317,860	258,655	667	2,173,390	99,244	289
1,317,860	272,680	682	2,173,420	72,574	289
1,317,860	306,910	700	2,174,500	65,144	289
1,317,860	307,276	719	2,219,320	59,621	454
1,317,860	335,429	751	2,222,490	52,654	495
1,317,860	359,092	777	2,223,640	46,194	321
1,318,560	441,777	883	2,813,060	40,074	364
1,319,060	138,811	366	2,814,070	32,719	470
1,319,350	362,124	902	2,814,150	35,184	262
1,321,730	469,515	959	3,482,140	30,184	244
1,322,520	459,399	978	3,482,140	28,935	244
1,323,320	360,443	997	4,642,770	26,184	417
1,324,190	429,769	1018	4,642,770	25,935	445
1,363,860	133,902	512	4,642,770	25,735	452

. Our analysis of these solutions reveals significant differences in the decision variables that drive performance across the three objectives.

The minimum energy consumption solution (1,317,860 GJ/year) achieves a 75.5% reduction in primary energy use compared to the baseline through a balanced investment across multiple building components. This solution selects high-performance building envelope components, with double glazing 4-12-4 argon-filled windows (U-value: 1.6 W/m²K) replacing standard double glazing, single-glazed flat surface solid doors (17% glass, U-value: 2.1 W/m²K) replacing double wing photocell doors, and external wall insulation thickness increasing from 5 cm to 8 cm (stone wool). Likewise, the most efficient coverings were chosen for the ceiling and floor. For heating systems, a high-efficiency condensing boiler (85% efficiency) coupled with improved control systems is selected. This solution also replaces all lighting with high-efficiency LED systems and upgrades to energy-efficient appliances with the highest available ratings. By directing all investment toward reducing the building’s intrinsic energy demand rather than generating renewable energy or supporting electric vehicles, this configuration minimizes total energy consumption from all sources.

The minimum investment solution ($25,735) employs a targeted approach that strategically allocates resources to the most cost-effective interventions. To optimize economic efficiency, the model retains the existing door and window systems while forgoing additional thermal envelope treatments for walls, ceiling, and floor surfaces. Given the absence of a hot water system in the existing building, the model selects the most economical non-electrical domestic hot water system. Furthermore, due to the significant contribution of electricity costs to the annual energy consumption calculation and the prevailing high electricity tariffs, the optimization framework determines an optimal configuration of 33 high-capacity (550 W) photovoltaic panels installed on the roof surface, coupled with a 30 kW inverter to facilitate efficient energy conversion and utilization.

In contrast, the maximum sustainability solution (1034) prioritizes sustainability scoring over pure energy efficiency. This solution installs the maximum number of solar panels in the available area (775 panels to the roof and 411 panels to park, primarily 375 W models), requiring additional structural investments. It incorporates a comprehensive EV charging infrastructure with 22 charging stations capable of supporting 50 electric vehicles. While building envelope improvements are similar to the minimum energy solution, this configuration dedicates a significantly higher portion of the investment to RES and transportation electrification, directly targeting the transportation metrics in the UI GreenMetric framework.

The balanced solution (1,367,010 GJ/year, $127,064, and 538 points) demonstrates a balanced trade-off across all three competing objectives. This configuration implements moderately high-performance building envelope components while maintaining the existing fenestration systems with their characteristic thermal transmittance value, complemented by an efficient non-electrical heating–DHW system (oil-based condensing combi boiler). The renewable energy component comprises 27 high-capacity (550 W) photovoltaic modules installed on the roof surface, coupled with a 30 kW inverter to maximize energy conversion efficiency and system performance. Notably, this balanced solution does not incorporate electric vehicle integration into the campus transportation infrastructure, consequently eliminating the requirement for charging station installation. This configuration thus represents an intermediate implementation strategy that achieves meaningful sustainability improvements while maintaining reasonable investment parameters.

Our analysis of the Pareto front in Table 5 reveals several critical trade-offs. First, achieving minimum energy consumption (1,317,860 GJ/year) requires moderate to high investment ($146,271–$359,092) but yields relatively low sustainability scores (366–777 points), highlighting the divergence between pure energy efficiency and broader sustainability metrics. Second, maximum sustainability scores (>900 points) require substantially higher investments ($360,443–$485,674) while accepting slight increases in energy consumption (1.4–4.8% above the minimum), demonstrating that transportation electrification comes at both financial and energy costs. Third, solutions with similar energy consumption or investment levels can have dramatically different sustainability scores, indicating that resource allocation strategies (whether directed toward building envelope improvements, renewable generation, or transportation electrification) significantly impact sustainability outcomes despite similar financial commitments. Finally, the balanced solution ($127,064) demonstrates that with strategic allocation—focusing on moderate envelope improvements and renewable energy while deferring transportation electrification—institutions can achieve near-optimal energy performance (only 3.7% above minimum) with substantial sustainability benefits at approximately 73% less investment than the maximum sustainability configuration.

Our analysis of decision variable patterns across the solution space reveals that as investment constraints tighten, the model prioritizes envelope improvements and energy system upgrades before renewable generation, as these typically offer higher energy savings per dollar invested. EV infrastructure is generally the last priority from an energy perspective but becomes increasingly important when sustainability scoring is weighted heavily. Higher-investment solutions typically select larger capacity components (550 W panels, 30 kW inverters) for space efficiency. These patterns provide valuable insights for phased implementation approaches, allowing institutions to strategically sequence investments according to available resources while maximizing performance across all objectives.

These results demonstrate the efficacy of an integrated approach to campus energy system optimization. Through concurrent consideration of building energy performance, renewable generation capacity, and transportation electrification, the model identifies configurations that synergistically enhance sustainability metrics while preserving economic feasibility.

5. Conclusions

The multi-objective optimization approach employed in this study yielded a diverse set of Pareto-optimal solutions for campus energy systems. Analysis of these solutions reveals distinct configurations that represent different trade-offs between energy consumption, investment costs, and sustainability metrics.

The minimum energy consumption solution ($146,271 investment) achieves the lowest possible energy footprint (1,317,860 GJ/year) while attaining a sustainability score of 366 points. This configuration demonstrates that significant energy efficiency improvements can be realized with moderate capital expenditure, offering an attractive option for institutions prioritizing operational energy reduction.

For budget-constrained institutions, the most economically viable implementation ($25,735) provides 4,642,770 GJ/year of energy consumption and 452 sustainability points. This configuration represents an accessible entry point into campus sustainability initiatives with minimal upfront investment while still delivering measurable environmental benefits.

The maximum sustainability score configuration ($485,674) reaches 1034 points through comprehensive renewable energy integration and complete transportation electrification. While requiring the highest investment and accepting a moderate increase in energy consumption (1,764,160 GJ/year), this solution would position the institution at the forefront of campus sustainability rankings.

A practical implementation is the balanced solution ($127,064), which achieves minimum energy consumption (1,367,010 GJ/year) while securing a substantial sustainability score of 538 points. This configuration effectively balances all three competing objectives, offering institutional decision-makers a robust implementation pathway that delivers strong performance across all metrics.

These results demonstrate that the inherent trade-offs in multi-objective optimization cannot be eliminated but can be strategically managed through informed decision-making. The optimization framework developed in this study provides campus administrators with a comprehensive tool for identifying configurations that align with their specific institutional priorities, budgetary constraints, and sustainability commitments. The findings further indicate that significant sustainability improvements are achievable at intermediate investment levels, making meaningful environmental progress accessible to a broad spectrum of higher education institutions.

Trade-offs between competing objectives were quantified through a comparative analysis of key Pareto solutions. The balanced solution ($127,064) achieves nearly optimal energy performance (1,367,010 GJ/year, just 3.7% above minimum) while requiring 73.8% less investment than the maximum sustainability configuration. Similarly, this balanced approach secures a substantial sustainability score (538.3 points) at only 26.2% of the investment required for the maximum score solution. The minimum cost solution ($25,735) demonstrates that even with limited resources, meaningful sustainability improvements are achievable (452.2 points), though with significantly higher energy consumption (4,642,770 GJ/year). These quantified correlations between capital expenditure and performance metrics across multiple objectives provide campus administrators with concrete metrics for evaluating implementation alternatives according to their specific institutional priorities and budgetary constraints.

While the optimization framework delivers valuable insights for campus energy planning, several implementation challenges should be considered when translating these theoretical results into practical applications. First, the administrative complexity of coordinating multiple stakeholders across campus departments (facilities management, transportation services, and sustainability offices) may introduce delays in the adoption of comprehensive solutions. Second, the phased implementation of recommended configurations requires careful sequencing to maintain system integrity while component upgrades occur at different times. Third, the integration with existing building management systems presents technical hurdles that vary by campus, potentially requiring customized interfaces between optimization outputs and operational controls. Additionally, market volatility affecting technology costs and shifting regulatory landscapes regarding renewable energy policies might necessitate periodic recalibration of the optimization parameters. Future research should address these practical deployment challenges through case studies of real-world implementations, the development of change management protocols specific to campus environments, and the creation of flexible implementation roadmaps that can adapt to institutional constraints while maintaining progress toward sustainability objectives.