Sage Journals: Discover world-class research

Abstract

Retrofitting existing buildings is widely regarded as a key strategy for minimising the environmental impact associated with buildings throughout their life cycle. Although significant attention has been given to residential buildings and effective measures have been implemented, non-domestic buildings still require focused efforts to meet the UK’s sustainability targets. The healthcare sector is a major contributor to the existing non-domestic building stock due to its energy-intensive nature. In the UK, the National Health Service (NHS) operates a wide variety of hospital types, making it difficult to develop effective retrofit strategies that are suitable for all. Therefore, a strategic approach to retrofitting that is tailored to the specific needs and preferences of each NHS Trust is essential. This study examines a range of retrofitting measures for deep-plan and tower Hospitals (DPTH) and identifies the optimal design for reducing life cycle carbon emissions while accounting for stakeholder priorities. The results suggest that while certain optimal scenarios are effective in reducing operational energy use, further life cycle assessment (LCA) reveals different scenarios with a greater impact on lowering the building’s life cycle carbon emissions. Hence, additional life cycle carbon assessment is necessary to evaluate the full impacts of the optimal retrofit designs.

Practical Application

This research offers building services engineers and sustainability consultants with a robust methodology to evaluate HVAC and glazing retrofit strategies for the energy-intensive areas in UK hospitals, aiming to minimise life cycle carbon emissions. By integrating stakeholder priorities into multi-objective optimisation study through multi-criteria decision-making analysis, the research helps practitioners identify retrofit scenarios that deliver significant carbon savings beyond operational energy reductions, supporting the development of more sustainable healthcare buildings.

Keywords

Building retrofit multi-objective optimisation decision-making life cycle analysis

Introduction

In respect to climate change and GHG emissions, buildings have a significant potential for the cost-effective reductions. Therefore, future reductions of buildings’ emissions play an important role to reduce the overall UK emissions.¹ In the 1960s, sustainable development started to be a major international concern due to reduction in natural resources and high energy consumption levels, and a variety of law and policies focused on the environmental protection as a result of the sustainability concerns.² It is emphasised that there is an absolute need to increase the sustainable performance of buildings and the construction industry.³ Therefore, various strategies such as adaptation and mitigation of the building sector, using renewable energy, refurbishment of existing building stock and multi-optimisation approaches have been suggested to improve the sustainability of the built environment.^4–7

Several studies indicated that existing buildings is a major source of energy saving potential,^8–11 therefore, the existing building stock is needed to be more energy efficient and environmentally friendly. In the UK, the non-domestic building stock, primarily hospitals and schools, is responsible for approximately 5% of total energy consumption, while all non-domestic stock accounts for around 10%.¹² A recent report¹³ has highlighted a shift in trends, with non-domestic emissions increasing by 1% in 2018, yet residential emissions falling by 2%. The health sector is a prominent contributor to the non-domestic building stock. Given the higher level of operating activity of hospital buildings in comparison to other non-domestic sectors, a greater consumption of energy and subsequent emissions are to be observed. In England, the National Health Service (NHS) is responsible for the production of 4% of the total carbon footprint.¹⁴ Since 2008, the NHS has been monitoring and reporting its carbon footprint, continuously optimising its methodologies to align with the requirements of the Climate Change Act.¹⁴ Despite these efforts, healthcare buildings in the UK remain energy intensive,¹⁵ and 4.6% of the UK’s total carbon emissions, equivalent to 22.8 MtCO₂e, linked to the NHS.¹⁶

The NHS estate consist of a diverse mix of building types constructed over the last century,¹⁷ with most facilities dating from the 1950s to the 1980s. Each NHS Trust manages a variety of hospital buildings serving different functions, thereby complicating the implementation of standardised energy efficiency measures across the site. Pavilion, deep plan, tower, low-rise, modular and courtyard buildings are the primary hospital typologies in the UK. Among these, deep-plan and tower hospitals are typically associated with higher energy consumption,^18,19 largely due to their complex layouts and the substantial energy demands required for their operation. In order to achieve a successful energy and carbon reduction levels for these hospitals, the key requirements have been identified as refurbishment and the implementation of optimal design solutions.²⁰ Nevertheless, the decision-making process that informs the selection of building refurbishment measures is complex and influenced by multiple factors, including building type, occupancy, cost and sustainability.²¹ In addition to these factors, there are trade-offs between different measures in the selection process of refurbishment measures for each building. Thus, a multifaceted approach is paramount to deliver sustainable retrofitting solutions in existing healthcare buildings.

This paper aims to provide a methodology that combines multi-objective optimisation (MOO), life cycle assessment (LCA), and a multi-criteria decision-making (MCDM) approach to develop informed retrofitting strategies for deep plan/tower hospital (DPTH) buildings in the UK. In this framework, MCDM was first applied to identify and prioritise the key retrofit objectives based on stakeholder preferences which informed the MOO process. Subsequently, the trade-offs among key performance objectives (e.g. operational energy use and cost) were explored through MOO analyses using building performance simulations. The resulting Pareto-optimal solutions were then evaluated using a comparative LCA to assess the life cycle carbon impacts that were not fully captured by the simulation-based optimisation. This integrated approach was used for the identification of retrofit strategies that balance performance, cost, and environmental impact in a representative hospital case study.

Literature review

Building retrofitting strategies

There are various approaches to understand the economic, social and environmental impacts of buildings and strategies to improve the efficiency in the current body of knowledge.^22–25 Retrofit and replacement options are the most accentuated strategies for existing building stock. It is stated that the research areas regarding retrofitting and sustainable building strategies are typically energy related.²⁶ Nevertheless, energy and carbon emissions are not the only concerns when considering the environmental impacts of retrofitting, but also the waste management, production of building materials, emissions from greenhouse gases (GHGs) are the other significant environmental concerns must be focused.²⁷ There is a growing number of studies suggesting the extension of lifecycle of buildings through retrofitting, which is much preferable to demolition in terms of improved economic, environmental and social impacts as well as energy.^7,27–29

The retrofitting of six public buildings was investigated in a study,²⁸ which concluded that improving the thermal insulation of the building envelope was the most significant measure for reducing energy consumption and CO₂ emissions. A further study evaluated a broader range of retrofit actions, including envelope insulation improvements, the installation of more efficient boilers, the use of photovoltaic systems and the consideration of the embodied energy of building materials.³⁰ This study concluded that an approximate 80% reduction in cumulative energy demand (CED) was achieved when these measures were implemented in combination, highlighting the substantial impact of integrated retrofit strategies.

In the literature, studies also focused on different aspects of LCA regarding building retrofit. An extended review on LCA of building refurbishment revealed that only a limited number of studies have comprehensively analysed retrofit processes and their environmental impacts.²³ To design and evaluate energy demand retrofitting scenarios, an approach was presented with the assessment of long-term cost effectiveness.³¹ This study associated various approaches, and life cycle cost analysis, energy demand modelling and retrofit option rankings were used for evaluation. Results showed that retrofit actions can achieve significant energy savings but are not always cost-effective.

Given the growing interest in the building performance modelling and optimisation, recent studies applied several approaches in order to improve the performance of the existing buildings through retrofitting. A review study indicated that approximately 20-30% reduction of energy consumption is an achievable target through building optimisation,³² nevertheless, further investigation on different approaches is needed due to variety of design objectives and limited results. Over the past decade, most research has focused on optimisation at the early design stage; however, extending these approaches to later design stages may also provide valuable insights.³³

In real-building environment, there are more than one design criteria (e.g. building operations, energy consumption, cost) that need to be addressed simultaneously in some cases. Hence, the selection of building retrofitting strategies involves a complex decision-making process.¹⁹ Therefore, using multi-objective optimisation approach might be more relevant than single objective,³² as it allows different performance targets to be evaluated together. Multi-criteria decision-making (MCDM), on the other hand, is another useful approach to problem solving that involves breaking down complex decisions into smaller and manageable components. While these methods provide valuable insights, further research is needed on the decision-making process incorporating with building performance optimisation in order to find the optimal refurbishment strategy to improve both energy efficiency and life cycle carbon.

NHS estates and sustainability

There was a considerable demand for hospital care at the time of the establishment of the NHS.³⁴ The NHS has its own structure and organisation and the medical care is divided into three levels of service: Primary Care, including community care, general practices (GP), pharmacists, dentists etc., Secondary Care, comprising acute hospitals accessed through GP referral, and Tertiary Care, which includes specialist hospitals.³⁵ The NHS building stock is diverse in terms of age, with the majority of structures having been constructed during the 1960s and 1970s. A substantial proportion of these buildings are currently at risk due to their age and require deep retrofitting.

To achieve the UK target for an 80% reduction in buildings’ GHG emissions, the NHS has a target that all buildings to be “low carbon”,³⁶ referring to buildings designed and operated to minimise both operational and embodied carbon emissions. In other words, all new NHS buildings and major upgrades must comply with the NHS Net Zero Building Standard,³⁷ which requires significant reductions in both operational and embodied carbon emissions. Despite the noteworthy progress in reducing carbon emissions, there is still a significant challenge for achieving a net zero NHS.¹⁴ Therefore, targeting reductions in building energy consumption, which accounts for 10% of NHS carbon emissions as shown in Figure 1., is fundamental.

Figure 1.

Sources of NHS carbon emissions by activity type (adapted from¹⁴).

Hospitals are complex buildings occupied 24 hours a day and have unique and high energy requirements.³⁸ In hospital buildings, a huge amount of energy is needed, particularly for heating/cooling, lighting, ventilation, and other equipment, due to the large number of users. The characteristics of hospital buildings are therefore based on this high energy requirements. Moreover, it is emphasised that the building types dominate the long-term profits of retrofit measures.³¹ Department of Health, Estate and Facilities Division identified the key factors that have contribution to the increase in total energy consumption of NHS hospitals¹⁵ as follows:

(1) Enhanced use of comfort cooling to offset the heatwave effects

(2) Constructing more deep-plan facilities

(3) Large number of single rooms in order to reduce healthcare-acquired infection (HAI)

(4) Overlooked expenditures on the health-care environment, such as expenditure for reducing energy consumption.

Hospitals constructed in between 1960s-80s, were found to have around 30 – 40% higher annual energy costs than earlier-built hospitals.¹⁸ The deep plan and tower block design, a variation of the corridor ward design and primarily constructed in the mid-1970s,³⁹ is associated with high electricity consumption per sqm, with total annual consumption ranging from approximately 13,000 to 35,000 MWh.⁴⁰ Regardless of their type, deep-plan hospitals demonstrated higher electricity use due to the need to ventilate the internal core spaces using mechanical systems.¹⁸ Tower hospital building design, on the other hand, enables to have one complete building operation on a restricted site, and this concept solved the challenge of the multiplicity of services.^17,41 However, deep plan/tower buildings has been described as one of the key barriers to implement low-energy measures in hospitals by the Department of Health.¹⁷

Methodology

To define the retrofit strategies and identify the optimal design option for the case study DPTH building, this study aims to apply a mixed-method approach, which provides an integration and collection of both qualitative and quantitative data. To achieve this, a multi-criteria decision-making method was used to determine the retrofit design strategies, which were then incorporated into the simulation-based multi-objective optimisation analysis, where the optimal strategies were identified. Finally, a life cycle carbon analysis was conducted to evaluate the optimal design strategies in terms of their effectiveness in reducing energy consumption and carbon emissions.

Multi-criteria decision making

MCDM is a method that helps people in making decisions according to their preferences when there are multiple criteria in conflict.⁴² Among several MCDM methods available in the literature, the most commonly applied deterministic approaches include the Weighted Sum Model (WSM), Weighted Product Model (WPM), Elimination and Choice Translating Reality (ELECTRE), Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), and Analytic Hierarchy Process (AHP).⁴³ The Analytical Hierarchy Process (AHP) is a structured decision-making technique that incorporates both rational analysis and expert judgment to identify the most suitable option among multiple alternatives evaluated against a set of criteria.⁴⁴ Since the decision-making problems often involve numerous complex elements in the construction sector, AHP is an established approach due to its simplicity and user-friendly structure that validates subjective judgements and provide high level of consistency.⁴⁵ The application of AHP in the construction industry typically involves three main steps: developing a hierarchical structure, conducting pairwise comparisons, and verifying the consistency of the judgments.

In this study, an AHP approach was applied prior to the optimisation and life cycle assessment process. This method helped to define retrofit solutions based on the stakeholder preferences on retrofitting the DPTH building. The NHS Estates and Facilities (EF) team, responsible for building operation and management, took part in the AHP decision-making exercise. The AHP process and full outcome are thoroughly explained in recent paper.¹⁹ The summary of the decision-making analysis results that guided the simulation-based multi-objective optimisation of the case study, presented in Figure 2 –4 Below.

Figure 2.

Stakeholder preference on retrofitting hospital zones (adapted from¹⁹).

Figure 3.

Stakeholder preference on building envelope strategies¹⁹.

Figure 4.

Stakeholder retrofitting preference on building systems strategies¹⁹.

As part of the AHP exercise,¹⁹ stakeholders were provided with a selection of hierarchy criteria to rank. The main three categories consist of selection of retrofit measures, sustainability targets and hospital zones. The criteria for selecting retrofitting measures were grouped into two sub-categories: building envelope measures (Figure 3) and building systems measures (Figure 4). As both were considered equally important, they were each included in the optimisation process. Given the outcome of the ranking exercise and consistency analysis, windows and HVAC design strategies were prioritised to be applied into baseline model of the case study hospital. As illustrated in Figure 2, the preferred areas in the hospital were the operating theatres (OT) and intensive care unit (ICU); hence, the optimisation was conducted for these areas. In regard to the sustainability targets category, stakeholders ranked the criteria they considered most significant for the retrofitting of the building. As a result, reductions in energy consumption, cost, and complying with regulations and standards were identified as the primary objectives for the following optimisation analyses.

Multi-objective optimisation approach

This study applied a simulation-based multi-objective optimisation approach informed by the AHP decision-making outputs to evaluate retrofit strategies for hospital buildings. The integrated AHP outputs ensure that stakeholder preferences systematically inform the optimisation process. Consequently, the generated Pareto-optimal set of retrofit solutions enables decision-makers to assess trade-offs between competing criteria and develop strategies that best align with their key objectives. This approach builds on prior optimisation studies that focused on non-domestic buildings,⁴⁶ by explicitly accounting for constraints specific to hospital operations and objectives derived from stakeholder engagement.

Case study hospital building

The case study hospital is a typical Deep Plan Tower Hospital (DPTH), constructed in the late 1960s, featuring a deep-plan base structure with a 14-storey tower block above. There has been no major retrofitting carried out on the building, but only minor updates to internal finishes and the facade. In particular, the HVAC systems remain in their original condition and need a thorough evaluation.

For this study, actual energy consumption data, presented in Table 1., was obtained from the NHS EF team to analyse the building’s current performance and develop a baseline energy model. When compared to the average energy consumption of deep-plan and tower hospital types as reported in the Estates Return Information Collection (ERIC) 2018/19 dataset,⁴⁰ a marginally higher usage was observed for the case study building. Since the actual energy usage data still aligns with the deep-plan hospital average, the case study building is classified as a Deep Plan Tower Hospital (DPTH) type.²¹

Table 1.

The real building energy usage metered data.

Annual consumption	kWh	kWh/m2	kgCO2e	kgCO2e/m2
Electricity	6,131,899	137.30	1,275,435	28.56
Steam boiler	18,552,300	415.42	3,246,652	72.70
Total	24,684,199	553	4,552,087	101.26

Building optimisation

Building performance can be improved by the parametric simulation method; however, this approach is stated as time consuming due to the complex interactions of input variables on the final results. Instead, the computer building model is suggested to lessen the time consumed and provide better approximations through such methods as simulation-based or numerical optimisation.^32,47 Thus, in this study, DesignBuilder v7.3.1.3,⁴⁸ a user-friendly dynamic energy modelling software that uses the EnergyPlus simulation engine⁴⁹ to calculate performance data, is used to generate the simplified building energy model and optimal retrofit scenarios. DesignBuilder optimisation tool using a Genetic Algorithm (GA) based on the NSGA-II method introduced by Deb et al.⁵⁰ has been employed to identify the optimal scenario by analysing both stakeholder-led preferences and regulatory-based variables. NGSA-II is widely applied in building retrofit MOO research because of ensuring a higher population diversity and reliable Pareto front solutions by improving convergence performance in multi-objective optimisation problems.⁵¹ It also allows reliable trade-offs between conflicting objectives such as cost and energy usage. The trade-off front solutions generated during the optimisation process illustrates Pareto-optimal design configurations balanced based on preferred objectives. Thus, this method enables our research to systematically explore multiple design options and select solutions that achieve balanced scenarios.

Given the large-scale nature of the case study hospital building, simplification is a rational decision to ensure the feasibility of the analyses as complicated models with multiple zones and systems have the potential to severely delay the optimisation process.³² For this reason, simplified-yet-representative building energy models were developed, while maintaining accuracy to the characteristics of case study hospitals and to avoid complexity that could result in prolonged simulation times and convergence issues. The population size and number of generations for convergence were set to 10 and 30, respectively. To reduce computing time, these settings were kept at moderate levels for this study. Although this may result in some design combinations not being fully explored, such settings are recommended for complex models to achieve a balanced and computationally efficient evaluation. Besides, the 0.1 mutation rate and 0.9 crossover rate settings allow the distinct combinations to be discovered and avoiding clustering the similar design strategies. An annual simulation was conducted to capture the full-year impact of the selected design variables and retrofit scenarios. The optimisation analysis took about 52 hours to process. In total of 179 design strategies were generated with ten optimal scenarios.

Based on the stakeholder’s priorities on retrofitting, cost and energy consumption were considered as the optimisation objectives, which were set as net site energy consumption and construction cost in the model. Accordingly, building construction cost (Capex) - The basic cost of the building and site construction was selected as an optimisation objective because it directly reflects the upfront investment required for retrofitting measures and can be reliably quantified across scenarios. Following a thorough evaluation of the approximate costs of the design variables, it was determined that the library database of DesignBuilder, which were derived from a range of sources, including consultations with a professional cost modeller and published materials, was within an acceptable range for UK-specific cost calculations. Consequently, the original cost setting was maintained. Similarly, net site energy consumption (total building energy use) was chosen to represent the priority of reducing energy demand for stakeholders, as it provides a measurable metric of actual building energy use, allowing direct comparison between baseline and retrofit scenarios and provides a practical indicator of overall energy efficiency improvements. Besides, ICU and OT zones preferred by the stakeholders were included as targeted areas in the optimisation analysis. Table 2 presents the overall zones analysed and their glazing and floor area for further assessment in the life cycle carbon assessment.

Table 2.

Hospital zones analysed in the optimisation study.

Floor	Zones	Names	Glazing area (m2)	Floor area (m2)
1st floor	1	Oper theatre	0	121.67
2nd floor	4	Oper theatres 1	17.584	132.62
		Oper theatres north block	0	362.01
		Recover rooms	64.914	310.64
		Recover rooms, nurse station	10.956	85.32
3rd floor	6	1 ICU bed ward	33.404	321.31
		1 ICU trolley	3.099	140.75
		1 recovery room	26.752	147.83
		1 theatres 2	0	494.79
		1 theatres, anasthetic rooms	45.151	238.1
		1 theatres, anast. Rooms, stores	0	552.78
		TOTAL	201.860	2907.82

According to the AHP study¹⁹ conducted prior to the optimisation process, the design variables were identified as HVAC, glazing and shading based on stakeholder priorities. The decisions on HVAC and glazing options were guided by current UK building regulations and standards,^20,52,53 which is another priority for stakeholders when considering retrofitting strategies. Therefore, U-values ranging from 1.3 W/m²·K to 0.7 W/m²·K were applied to the glazing options. On the other hand, HVAC systems chosen for the optimisation analyses were representative configurations commonly used in UK hospitals. The simplified DesignBuilder templates were used to capture main operational characteristics without modelling detailed system complexities. For example, although the database also includes an air-source heat pump (ASHP) with convectors, this system was not selected as it incorporates natural ventilation alongside mechanical heating within the template, which is unsuitable for the ICU and OT zones. Instead, an ASHP hybrid with a gas boiler was selected to represent a typical forced-air configuration, allowing consistent evaluation across optimisation strategies. Presented in Table 3 below, a total of twenty-nine design variables were evaluated, with the ICU and OT zones identified as target objects for HVAC, and the window areas of these zones assigned as target objects for shading and glazing variables.

Table 3.

Design variables in optimisation study.

	Design variables	Number of options	Options	Target object
Building systems	HVAC	13	• Air to water heat pump (ASHP) hybrid with gas boiler	Intensive care unit (ICU) & operating theatre (OT)
			• CAV, air cooled chiller, 4-pipe induction units
			• CAV, water cooled chiller, boiler HW
			• Cooled beams, air cooled chiller, DOAS
			• Chilled ceiling, air cooled chiller
			• GSHP Unitary water-to-air heat pump
			• Fan coil unit (4-pipe), air cooled chiller, DOAS
			• PTHP
			• Unitary heat pump
			• VAV, air cooled chiller, reheat
			• VAV, dual duct, air cooled chiller
			• VAV, water cooled chiller, full humidity control
			• VRF (air cooled), heat recovery, DOAS, DCV
Building envelope	Glazing	12	• Triple clr 3 mm/13 mm Arg	Windows
			• Triple clr 6 mm/25 mm air for mid pane blinds
			• Tripple LoE (e2 = e5 = 1) clr 3 mm/13 mm air
			• Tripple LoE (e2 = e5 = 1) clr 3 mm/13 mm arg
			• Tripple LoE (e5 = 1) clr 3 mm/13 mm Arg
			• Tripple LoE film clr 6 mm/13 mm air
			• Dbl LoE spec sel clr 3 mm/13 mm/6 mm Arg
			• Dbl LoE Elec ref colored 6 mm/13 m Arg
			• Dbl LoE (e2 = .1) clr 3 mm/13 mm Arg
			• Dbl Elec ref bleached 6 mm/13 mm air
			• SGG PLT XN 4/16/4
			• SageGlass climaplus classic SR2.0 no tint
	Shading	4	• Louvre, 0.5 m projection + 0.5 m overhangs and sidefins	Windows
			• 1.0 m protection louvre
			• 1.0 m Overhang
			• No shading

Life cycle carbon analysis

Life Cycle Assessment, which is also known as life cycle analysis, evaluates the inputs, outputs and environmental impacts of a product system from cradle to grave throughout its life cycle.⁵⁴ The International Organisation of Standardisation (ISO) developed and released a methodological framework for LCA analysis with four phases: Goal and scope definition, Inventory analysis, Impact assessment and Interpretation. The main goal of this study was to evaluate the environmental impact of retrofitting ICU and OT departments of a DPTH building in the UK, focusing on the reduction of life cycle carbon emissions. This study adopted a cradle-to-grave system boundary from extraction of raw material to end-of-life management, aligns with BS EN 15,978 and RICS Whole Life Carbon Methodology.^55,56 For each retrofit scenario, the assessment included the product and manufacturing stages (A1–A3), which represent the embodied carbon associated with raw material extraction and transport to manufacturing, as well as use stage modules (B1–B2) where applicable (e.g. maintenance or minor replacement of components). The operational energy use (module B6) data was generated through DesignBuilder simulations and added into the LCA model. The conversion of the operational energy data imported into the One Click LCA platform was calculated using UK electricity and gas emission factors from the 2024 Government greenhouse gas conversion factors published by Department for Energy Security & Net Zero.⁵⁷ End-of-life stages (C1–C4) were also incorporated through the Environmental Product Declarations (EPDs) embedded within the One Click LCA database. The removal of baseline systems was not considered separately, since identical demolition and disposal processes would apply to all retrofit options. Consequently, their carbon impacts would become negligible in a comparative assessment. This simplification is consistent with BS EN 15,978 and RICS methodology, which allows the exclusion of life cycle modules that do not influence comparative results. The analysis excluded the beyond the life cycle (module D) emissions due to uncertainty in future reuse, recycling and associated impacts. Functional unit was defined as 1 m² of gross floor area of the ICU and OT zones over a 60-year life cycle period. In this study, the scope of the building components included is limited to those directly affected by the proposed retrofit measures within ICU and OT zones. The only components changed between scenarios were analysed in LCA, which are: HVAC systems, glazing options and shading as prioritised by stakeholders. The material details of the optimal retrofit scenarios analysed are given in Table 4.

Table 4.

Summary of optimal retrofit strategies.

		HVAC	Glazing	Shading
	Baseline	VAV, air cooled chiller, reheat	Double clear 3 mm/13 mm air	Blind with medium reflectivity (inside)

	1	ASHP hbrid with gas boiler	Triple LoE e5 = 1 clr 3 mm/13 mm argon	No shading
	2	ASHP hbrid with gas boiler	SGG PLT XN 4/16/4	No shading
	3	PTHP	Double LoE Ele ref coloured 6 mm/13 mm argon	0.5 m Louvre+0.5 m Overhang with sidefins
	4	PTHP	Double LoE Ele ref coloured 6 mm/13 mm argon	1m Louvre
Optimal design	5	PTHP	Double LoE Ele ref coloured 6 mm/13 mm argon	1m Overhang
	6	PTHP	Double LoE Ele ref coloured 6 mm/13 mm argon	No shading
	7	PTHP	Sageglass climaplus classic SR 2.0 No tint	0.5 m Louvre+0.5 m Overhang with sidefins
	8	PTHP	Sageglass climaplus classic SR 2.0 No tint	1m Louvre
	9	PTHP	Sageglass climaplus classic SR 2.0 No tint	1m Overhang
	10	PTHP	Sageglass climaplus classic SR 2.0 No tint	No shading

For the LCA analysis, One Click LCA tool⁵⁸, sustainability platform that supports decarbonisation in construction and manufacturing, was used. For the life cycle inventory data, the platform uses public EPD sources as well as licensed LCI databases i.e. Ecoinvent v3.10.1,⁵⁹ and its own internal LCI datasets. The CML baseline method was employed in this study. The EN 15,804 standard, which is in line with the ISO 14,044/44 standard, mainly refers to the application of CML as it is a commonly used life cycle impact assessment (LCIA) method for construction.⁶⁰

Firstly, a baseline scenario for life cycle carbon evaluation was created in the LCA tool using the input data from the energy model. The baseline carbon model was developed particularly for HVAC, glazing and shading designs, in accordance with the scope of the life cycle assessment, as discussed above. The operational energy data for optimal retrofit strategies were gathered from the previous optimisation analyses and included into the LCA tool energy consumption module (B6). Steam boiler and electricity were used as fuel types, and all emissions were calculated using the relevant UK emission factors as 0.175 kgCO₂/kWh for steam boiler and 0.208 kgCO₂/kWh for electricity. The conversion factors were taken from the 2024 Government greenhouse gas conversion factors published by Department for Energy Security & Net Zero.⁵⁷

A total of five optimal retrofit design strategies were assessed, and the results were compared with a baseline scenario that represents the as-built condition with existing HVAC and double-glazed windows. The assessment focused on global warming potential (GWP) over 60-year lifespan.

Results

The first part of this section discusses the AHP-informed MOO results for the case study hospital, followed by life cycle assessment of the optimal strategies to determine a retrofit scenario for ICU and OT zones that achieves the greatest carbon reductions. Lastly, a comparison between the energy and carbon results of the optimal design options is conducted. Through trade-off analysis, the most balanced and suitable retrofit scenario for the case study building is identified.

Multi objective optimisation results

As discussed in detail in section 3.2, the scope of multi-objective optimisation, also known as Pareto optimisation, is to investigate the trade-offs among multiple conflicting objectives. The pareto front graph generated as a result of the optimisation analysis is shown in Figure 5. The lowest figures on net site energy consumption were recorded at 21.1 M kWh, with an associated cost of approximately £84.2 million. This optimal design is part of the 5^th generation of the optimisation process and consists of Triple Low-e 3 mm/13 mm glazing, ASHP with no shading. The first optimal scenario, however, recorded on the 3^rd generation, and it is the least costly scenario among other nine optimal scenarios. Whilst the cost values produced by optimisation engine reflect full-building construction estimates rather than incremental retrofit investments, the model was used comparatively to assess the relative differences between retrofit scenarios. Since the same cost database and assumptions were applied across all cases, the relative ranking of cost and energy performance remains valid for identifying the most balanced retrofit strategy.

Figure 5.

Pareto front graph of the building optimisation.

The Pareto front displayed a trade-off pattern between energy consumption and cost. Although the application of more advanced retrofit measures can improve energy performance, the potential cost savings may be low. This kind of behaviour is characteristic of multi-objective optimisation in complex building retrofits. In such cases, mechanical system upgrades typically dominate the energy benefits yet simultaneously represent the highest capital investments.

The baseline energy model has variable air volume (VAV) mechanical system, double glazing and inside blinds with medium reflectivity according to the collected building information. Details of the ten optimal design strategies are demonstrated in Table 4. In terms of energy efficiency, air source heat pumps (ASHPs) are highly efficient and a great option to reduce operational carbon emissions. The packaged terminal heat pump (PTHP) system, on the other hand, is known to have individual unit-level control, which may help to maintain zone-specific controls. Thus, both mechanical systems could provide better system management as well as lowering operational energy consumption.³⁷

While the ASHP and PTHP systems dominates the mechanical intervention, a variety of glazing options is observed as part of the optimal strategies. Besides, nearly half of the optimal shading design strategies include no shading, likely due to the north-facing orientation of the ICU and OT zones, which receive minimal direct solar exposure. The findings suggest that when ASHP is chosen as the HVAC system for ICU and OT zones, shading is not required. Nevertheless, no direct correlation was found between glazing type and shading variables.

Data presented in Table 5 shows the most effective optimal designs and the operational energy difference per square meter between baseline building and retrofitted building. A noteworthy outcome is that the savings are around 22 kWh/m² on site level when an ASHP system is applied, nevertheless, this figure drops significantly to approximately 8 kWh/m² in optimal scenarios with a PTHP system. Much larger reductions of up to 44% with ASHP and glazing upgrades were achieved for the ICU and OT zones, which are highly energy-intensive despite representing only around 8% of the total building area. These findings are comparable with the previous studies on hospital retrofits,⁶¹ in which the combination of HVAC, lighting and window upgrades resulted in approximately 48% reductions in high-demand areas. On the other hand, the difference between shading and glazing materials has a minimal impact compared to mechanical system replacement. Given these smaller differences, the five optimal designs presented in Table 5 were considered as representative optimal design scenarios for the next stage.

Table 5.

Building operational energy savings for each optimal retrofit design.

Optimal design	Baseline building (kWh/m2)	Retrofitted building (kWh/m2)	Savings (kWh/m2)
1	506.16	483.92	22.24
2	506.16	483.93	22.23
3	506.16	497.43	8.73
8	506.16	497.53	8.63
10	506.16	497.57	8.59

Although the 4% reduction in site energy per square metre may seem insignificant, it is indicative of the retrofit scope and the high operational intensity of critical hospital zones. Despite the fact that a conventional return-on-investment assessment may indicate a low immediate financial payback, such measures have been demonstrated to deliver broader co-benefits, including improved system reliability, reduced maintenance needs, improved occupant comfort, and alignment with NHS Net Zero targets. Considering the Cost-Benefit and Co-Benefit Analysis of Retrofitting (CAR), it is imperative to acknowledge the significance of non-financial benefits, particularly in the context of healthcare facilities where resilience and environmental quality are critical performance outcomes.

In this section, results highlight that optimal designs favouring ASHP configurations demonstrate the most substantial operational carbon savings and therefore require further evaluation through the life cycle carbon assessment stage. By contrast, scenarios involving PTHP systems may offer more precise temperature control and greater comfort in each room, while potentially offering lower overall energy savings. This could affect the trade-offs between embodied and operational carbon in the next LCA process.

Life cycle analysis results

The pre-retrofit baseline scenario for ICU and OT zones has total carbon emission of 26,102,552 kg CO₂e. The summary table below (Table 6) presents the life cycle carbon emissions of each representative optimal scenarios identified in the previous analysis. Despite achieving the largest reduction in operational energy, the ASHP combined with triple low-e glazing (Optimal Design 1) resulted in the smallest decrease in total carbon emissions, with a reduction of only 349,189 kg CO₂e. This contradictory outcome can be explained by the fact that the higher embodied carbon associated with the ASHP system and triple-glazed low-e windows. PTHP, on the other hand, is a decentralised, self-contained unit with lower material and installation requirements compared to ASHP system that typically involves larger compressors and extensive ductwork and refrigerants. For that reason, the combination of PTHP and double low-e glazing alongside with 0.5 m Louvre + 0.5 Overhang with sidefins (Optimal Design 3) achieved a greater reduction in life cycle carbon emissions at 537,210 kg CO₂e, despite lower operational energy savings. Besides, shading strategies, although varied across scenarios, demonstrated a smaller impact on total carbon. Meanwhile, Optimal Design 10 achieved the second-highest reduction in total carbon emissions despite offering the lowest operational energy savings. This further shows the trade-offs between operational and embodied carbon retrofit scenarios.

Table 6.

Life cycle carbon emission savings for optimal retrofit strategies.

Baseline Scenario (kg CO2e)	Optimal Scenario 1 (kg CO2e)	Optimal Scenario 2 (kg CO2e)	Optimal Scenario 3 (kg CO2e)	Optimal Scenario 8 (kg CO2e)	Optimal Scenario 10 (kg CO2e)
26,102,552	25,753,363	25,710,996	25,565,342	25,591,533	25,591,517
	−349,189	−391,556	−537,210	−511,019	−511,035

By integrating low-carbon mechanical systems and advanced glazing technologies, Optimal Design 3 demonstrates a retrofit strategy that not only meets energy and carbon objectives but also supports hospital resilience while also delivering economic, environmental, and social benefits.⁶² Overall, these results highlight the importance of considering both operational and embodied carbon in decision-making process for building retrofits, since achieving the highest operational savings does not necessarily result in the lowest total carbon impact.

Limitations and future work

This study applied an integrated AHP–MOO–LCA framework to evaluate and optimise retrofitting strategies for a hospital case study, focusing on HVAC, glazing, and shading systems within the ICU and OT departments. The selection of these options was guided by NHS Estates stakeholders through the AHP process, ensuring the analysis reflected real operational priorities. However, this scope necessarily excluded other potentially impactful measures such as lighting, building envelope insulation, and renewable generation systems. Future work should expand the optimisation framework to include these additional parameters which can deliver significant reductions in operational energy demand, cost and embodied carbon.⁶¹

Although this paper reports results for a single hospital typology, the same methodology is currently being applied to two additional NHS hospital buildings. These further studies will validate the applicability of the framework under different stakeholder priorities and operational goals, where Estates teams may prioritise factors such as cost, disruption or energy savings at different levels. This will help to assess the robustness and adaptability of the approach in different decision-making contexts within similar deep-plan hospital typologies.

Conclusion

Given the importance of retrofitting existing healthcare buildings to meet the UK’s sustainability targets, this study aimed to identify the most balanced retrofit strategy for a typical DPTH case study building, which is an energy-intensive building type. This was done by using simulation-based multi-objective optimisation method informed by analytical hierarchy process (AHP), a multi-criteria decision-making approach. The AHP-informed optimisation provided tailored decision variables that align with stakeholder priorities and site benefits. Reductions in both operational and embodied carbon emissions were targeted; therefore, the final approach was life cycle carbon calculations of the optimal design strategies. Although the optimal design solutions were successfully identified, the integration of three different methods introduces potential sources of error in informing the next stage. Having obtained the weights of the criteria prioritised by stakeholders, their judgements were evaluated for consistency using the methodology developed by Saaty.⁶³ The overall consistency ratio of the AHP results was calculated as 0.1, indicating an acceptable level of internal consistency among stakeholder decisions as discussed by Doguc et al.¹⁹ Additionally, validating the simulation model against actual building energy consumption data produced an error margin of 8.4%, which aligns with the ± 10% Normalised Mean Bias Error (NMBE) threshold specified in ASHRAE Guideline 14⁶⁴ for calibrated energy models. Since the LCA was conducted as a comparative analysis limited to the updated building elements, a separate error calculation was not performed at this stage. It is worth to note that One Click LCA’s databases and standard assumptions provide industry-standard estimates for these comparative assessments.

The final results revealed that Optimal Design 3, which includes a PTHP system, double low-e glazing, and 0.5 m louvre + 0.5 m overhangs and sidefins, achieves a 2% reduction in life cycle carbon emissions of the targeted area and ranks third best option in operational energy usage reduction. Notably, applying shading has minimal to no impact on total building energy consumption when the analysed area faces north and relies entirely on mechanical systems, such as in ICU and OT zones in hospitals.

Relying solely on simulation-based optimisation results may increase the risk of misleading outcomes due to the different limitations of optimisation tools. Given the results of this study, Optimal Design 1 was the best-performing solution in terms of reducing operational energy consumption prior to the application of life cycle carbon assessment. However, following the LCA analysis, it was observed that Optimal Design 1 achieved the lowest reduction in life cycle carbon emissions. While simulation-based optimisation remains a user-friendly and valuable approach, additional support from carbon assessment tools is necessary when aiming for comprehensive reduction in building carbon emissions.

Footnotes

Acknowledgments

This research has been funded by the Republic of Turkey Ministry of National Education. There is no conflict of interest to declare in the study. We also thank the NHS Estates and Facilities team for being part of this study.

Author contributions

Kubra Doguc: Writing – review & editing, Writing – original draft, Visualisation, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualisation. Teresa Domenech Aparisi: Supervision, Investigation. Rokia Raslan: Writing – review & editing, Supervision, Methodology, Investigation

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research has been funded by the Republic of Turkey Ministry of National Education

Disclosure statement

During the preparation of this work, ChatGPT-4o was used only to identify and correct grammatical errors in the paper. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Ethical considerations

This study was undertaken as a part of an ongoing PhD research: Optimising life cycle performance of hospital building retrofitting through low carbon measures: An integrated methodological framework. This research study is registered with UCL Data Protection (Z6364106/2021/11/25) and has been approved by the UCL BSEER IEDE Research Ethics Committee (Project ID: 20211111_IEDE_PGR_ETH).

ORCID iD

Kubra Doguc

Data Availability Statement

The data that has been used is confidential.*

References

CCC . The sixth carbon budget: the UK’s path to net zero. Comm Clim Chang 2020; 448.

Shelton

Stockholm Declaration Rio Declar ation . Max Planck Encycl Public Int Law [MPEPIL] (1992). https://opil.ouplaw.com/display/10.1093/law:epil/9780199231690/law-9780199231690-e1608?rskey=msBFLd&result=2&prd=OPIL.

Marjaba

Chidiac

. Sustainability and resiliency metrics for buildings - critical review. Build Environ 2016; 101: 116–125.

Bullen

. Adaptive reuse and sustainability of commercial buildings. Facilities 2007; 25: 20–31.

Felgueiras

Martins

Caetano

. Sustainability in buildings – a teaching approach. Energy Proc 2017; 107: 15–22.

Kreiner

Passer

Wallbaum

. A new systemic approach to improve the sustainability performance of office buildings in the early design stage. Energy Build 2015; 109: 385–396.

Cooper

Daly

, et al. Existing building retrofits: methodology and state-of-the-art. Energy Build 2012; 55: 889–902.

Hasegawa

. Analytical study of policy for improving the energy efficiency of existing buildings in European countries and the USA. J Environ Eng 2003; 556: 103–110.

Uihlein

Eder

. Policy options towards an energy efficient residential building stock in the EU-27. Energy Build 2010; 42: 791–798.

10.

Baek

C-H

Park

S-H

. Changes in renovation policies in the era of sustainability. Energy Build 2012; 47: 485–496.

11.

OECD/IEA . Towards sustainable use of building stock: Workshop report. 2004.

12.

PIA . Property industry alliance property data report 2016 (2017). https://www.bpf.org.uk/sites/default/files/resources/PIA-Property-Report-2016-final-for-web.pdf

13.

CCC. Reducing UK emissions . Progress Report to Parliament, 2019, vol 267.

14.

NHS England Improvement NHS . Delivering a ‘Net Zero. National Health Service, 2020.

15.

Short

Al-Maiyah

. Design strategy for low-energy ventilation and cooling of hospitals. Build Res Inf 2009; 37: 264–292.

16.

Nhs

SDU

. Carbon footprint update for NHS in England 2015. NHS Engl.

17.

Fifield

L-J

. Monitoring UK hospital building type performance. Loughborough University, 2016.

18.

Williams

Knight

Griffiths

. Hospital energy performance: new indicators for UK national health service estate. Build Serv Eng Technol 1999; 20: 9–12.

19.

Doguc

Domenech Aparisi

Raslan

. Towards net-zero for hospital estates: stakeholder-led refurbishment strategies. In: eceee 2024 summer study proceedings. European Council for an Energy Efficient Economy (ECEEE), pp. 561–566.

20.

Department of Health . EnCO2de 2015 - making energy work in healthcare. Part B: procurement and energy considerations for new and existing building facilities. Heal Tech Memo 07-02. 2015; 132: 29–37.

21.

Doguc

Domenech Aparisi

Raslan

. A multi-criteria decision-making approach to optimising hospital estate refurbishment In: CIBSE IBPSA-England Technical Symposium. Chartered Institution of Building Services Engineers. CIBSE, 2025, pp. 1–12.

22.

Schwartz

Raslan

Mumovic

. The life cycle carbon footprint of refurbished and new buildings – a systematic review of case studies. Renew Sustain Energy Rev 2018; 81: 231–241.

23.

Cabeza

Rincón

Vilariño

, et al. Life cycle assessment (LCA) and life cycle energy analysis (LCEA) of buildings and the building sector: a review. Renew Sustain Energy Rev 2014; 29: 394–416.

24.

Vilches

Garcia-Martinez

Sanchez-Montañes

. Life cycle assessment (LCA) of building refurbishment: a literature review. Energy Build 2017; 135: 286–301.

25.

Ramesh

Prakash

Shukla

. Life cycle energy analysis of buildings: an overview. Energy Build 2010; 42: 1592–1600.

26.

Gohardani

Björk

. Sustainable refurbishment in building technology. Smart Sustain Built Environ 2012; 1: 241–252.

27.

Bell

Chaytor

Crawford

, et al. Making Decisions on the Demolition or Refurbishment of Social Housing Key points for making decisions 2 . Improving energy performance and, UCL Policy Briefing 2014. https://www.ucl.ac.uk/engineering-exchange/sites/engineering-exchange/files/policy_brief_refurbishment_and_demolition_of_social_housing.pdf

28.

Ardente

Beccali

Cellura

, et al. Energy and environmental benefits in public buildings as a result of retrofit actions. Renew Sustain Energy Rev 2011; 15: 460–470.

29.

Ding

Ying

. Embodied and operating energy assessment of existing buildings – demolish or rebuild. Energy 2019; 182: 623–631.

30.

Beccali

Cellura

Fontana

, et al. Energy retrofit of a single-family house: life cycle net energy saving and environmental benefits. Renew Sustain Energy Rev 2013; 27: 283–293.

31.

Wang

Holmberg

. A methodology to assess energy-demand savings and cost effectiveness of retrofitting in existing Swedish residential buildings. Sustain Cities Soc 2015; 14: 254–266.

32.

Nguyen

Reiter

Rigo

. A review on simulation-based optimization methods applied to building performance analysis. Appl Energy 2014; 113: 1043–1058.

33.

Attia

Hamdy

O’Brien

, et al. Computational optimisation for zero energy buildings design: interviews results with twenty eight international experts. In: Proceedings of BS 2013: 13th Conference of the International Building Performance Simulation Association. 2013, pp. 3698–3705.

34.

Rivett

. From cradle to grave: fifty years of the NHS. Book, London: King’s Fund, 1998.

35.

Grosios

Gahan

Burbidge

. Overview of healthcare in the UK. EPMA J 2010; 1: 529–534.

36.

Nhs

SDU

. Carbon footprint update for NHS in England. Sustain Dev Unit 2013; 19.

37.

NHS England . NHS net zero building standard. 2022.

38.

Ziębik

Hoinka

. Energy systems of complex buildings. Book. London: Springer, 2013.

39.

Nazarian

Price

a DF

Demian

. A review of different approaches to access and people circulation within health-care facilities and the application of modelling, simulation and visualisation. Archit Manag Digit Arena 2011; 12.

40.

NHS Digital . Estates return information collection (ERIC) 2018/19 data quality report 2019 (2019). https://www.statisticsauthority.gov.uk/code-of-practice/the-code/

41.

James

Tatton-Brown

. Hospitals: design and development. Book, London: Architectural, 1986.

42.

Mardani

Jusoh

Nor

KMD

, et al. Multiple criteria decision-making techniques and their applications - a review of the literature from 2000 to 2014. Econ Res Istraz 2015; 28: 516–571.

43.

Triantaphyllou

. Multi-criteria decision making methods: a comparative Study/by evangelos Triantaphyllou. Dordrecht: Kluwer Academic Publishers, 2000.

44.

Saaty

. Models, methods, concepts & applications of the analytic hierarchy process/by Thomas L. Saaty and Luis G. Vargas. Boston; London: Kluwer Academic Publishers, 2001. https://www.loc.gov/catdir/enhancements/fy0819/00051460-t.html

45.

Darko

Chan

APC

Ameyaw

, et al. Review of application of analytic hierarchy process (AHP) in construction. Int J Constr Manag 2019; 19: 436–452.

46.

Pérez Téllez

Suárez Medina

León Rodríguez

ÁL

. Methodology for the optimisation of thermal performance and daylight access to the retrofit of hospital rooms in Mediterranean climate. In: Mercader-Moyano

(ed). Sustainable Development and Renovation in Architecture, Urbanism and Engineering. Cham: Springer International Publishing, pp. 403–413.

47.

Pan

Zhu

, et al. Building energy simulation and its application for building performance optimization: a review of methods, tools, and case studies. Adv Appl Energy 2023; 10: 100135.

48.

DesignBuilder . DesignBuilder software Ltd - Home. accessed 20 October 2025. https://designbuilder.co.uk/

49.

U.S. Department of Energy . EnergyPlus essentials. 2021.

50.

Deb

Pratap

Agarwal

, et al. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 2002; 6: 182–197.

51.

Costa-Carrapiço

Raslan

González

. A systematic review of genetic algorithm-based multi-objective optimisation for building retrofitting strategies towards energy efficiency. Energy Build 2020; 210: 109690.

52.

Department of Health . Estates and facilities divison. Health Building Note 00-01: General Design Guidance for Healthcare Buildings 2014; 77.

53.

HM Government . Approved document L, conservation of fuel and power, volume 2: buildings other than dwellings, 2021 edition incorporating 2023 amendments. Build Regul 2023: 23–25.

54.

ISO . Environmental management - life cycle assessment - principles and framework (ISO 14040:2006). ISO (International Organization for Standardization) (2006). https://bsol.bsigroup.com/Bibliographic/BibliographicInfoData/000000000030154435

55.

BS EN 15978:2011 sustainability of construction works. Assessment of environmental performance of buildings. Calculation method. BSI (2011). https://bsol.bsigroup.com/Bibliographic/BibliographicInfoData/000000000030256638

56.

RICS. Royal Institution of chartered surveyors professional statement: whole life carbon assessment for the built environment (2017). https://www.rics.org/

57.

Department for Energy Security and Net Zero . Government greenhouse gas conversion factors for company reporting: Methodology paper 2024: 1–54.

58.

One Click LCA, https://oneclicklca.com/en-gb/ (accessed 2 August 2025).

59.

Ecoinvent. Ecoinvent version 3.10.1 (2024, accessed 20 October 2025). https://support.ecoinvent.org/ecoinvent-version-3

60.

Guinée

. Handbook on life cycle assessment: operational guide to the ISO standards: Springer Science & Business Media, 2002.

61.

Dong

Hamada

Paydar

, et al. Identifying decarbonization pathways for primary healthcare buildings using energy certificate models in England and Wales. Build Environ; 274. DOI: 10.1016/j.buildenv.2025.112794, Epub ahead of print 2025.

62.

Vourdoubas

. Possibilities of using sustainable energy technologies including CHP systems, solar photovoltaics and heat pumps in hospitals. Eur J Energy Res 2022; 2: 1–8.

63.

Saaty

. In: Mitra

Greenberg

Lootsma

, et al. (eds). What is the Analytic Hierarchy Process? BT - Mathematical Models for Decision Support. Berlin, Heidelberg: Springer Berlin Heidelberg, 1988, pp. 109–121.

64.

ASHRAE . Guideline 14-2014: measurement of energy, demand, and water savings. American society of heating, refrigerating, and air conditioning engineers. Atlanta, Georgia, 2014, p. 2014.