Integrated life cycle assessment (ILCA) approach to daylighting renovation of a sports hall considering light quality, environmental and financial aspects Open Access

The building sector is a significant energy consumer, being responsible for 40% of global primary energy use and 55% of electricity use (Pérez-Lombard et al., 2008; Levermore, 2008; González-Torres et al., 2022), which corresponds to approximately 23% of the total CO₂ emissions coming from economic activities (Huang et al., 2018). At the European level, the building sector accounts for 36% of the greenhouse gas emissions (European Commission, 2021). While embodied carbon is particularly important for new buildings (Röck et al., 2020), in older constructions and in a life cycle perspective, the operational phase is generally the main energy consumer and carbon emission source (Gustavsson and Joelsson, 2010). In Sweden, sports facilities are particularly high energy consumers, using about 151 kWh/m² per year (The Swedish Energy Agency, 2010), significantly more than offices and schools (The Swedish Energy Agency, 2007a, b). Much of this – 31 kWh/m² – is attributed to lighting (Statisticon AB, 2018). Despite a strong reduction in electrical lighting use through more efficient technologies, much work remains to be done, with proper integration of daylighting providing the greatest opportunities for energy saving (Lundgren, 2016; Gentile et al., 2022; Gentile, 2022; Papinutto et al., 2022). Evidence suggests that lighting plays a key role in regulating physiological and psychological responses (Houser and Esposito, 2021), with access to daylighting being of particular importance for school children (Gentile et al., 2018). Also, daylight provides the right amount, timing and spectral composition of light for regulating physiological and psychological functions without necessarily increasing energy use for lighting (Taghizadeh et al., 2025). Light quality covers, therefore, vision and goes beyond that (Hosseini et al., 2024). Integrated lighting can combine daylighting, electric lighting and their controls to enhance visual performance, well-being and comfort, while reducing building energy use (Gentile et al., 2021, 2022). It is thus important to design lighting and daylight of the highest quality to see and feel well. However, designing or retrofitting for good daylighting could be costly (Wang et al., 2020; Scorpio et al., 2022), and the environmental impact of solutions might be sizeable (Eisazadeh et al., 2022, 2024; Rezaei Oghazi et al., 2024; Sällström and Lidelöw, 2024).

The environmental aspect is today of utmost importance, and Life Cycle Assessment (LCA) approaches are entering regulations. For example, since 2023, Denmark has introduced limits on CO₂ emissions for new buildings larger than 1,000 m² (Social-og Boligstyrelsen, 2024); since 2022, Sweden requires a climate declaration for all new buildings to obtain a building permit (Boverket, 2021). While a life cycle perspective is adopted in Nordic countries, even in Europe at large, the environmental aspect is coming into picture, with the latest Energy Performance for Building Directive (EPBD) envisioning zero-emissions for new and existing buildings by 2050 (European Parliament, 2024).

The aspects of light quality, environmental impact and cost taken together cover the three pillars of sustainable buildings as defined by the European standard framework (CEN, 2021b), with daylighting quality being expressly mentioned as supportive of social sustainability (CEN, 2014).

Light quality, environmental and financial requirements might lead to conflicting demands; it is important to adopt an integrated approach to the design decision process of daylighting projects (Schneider-Marin et al., 2022). A question, therefore, arises: how can decision-makers be supported in making more informed decisions about quality, environmental and financial aspects of daylighting projects?

To illustrate the scope, this study examines the renovation of an existing sports hall – currently lit exclusively by electric lighting – to incorporate daylight. It explores three options: tubular daylight devices (TDDs), skylights (SKL) and windows (SLW), assessed for daylight performance – including circadian potential – energy demand for lighting and heating, environmental impact through LCA, and costs via Life Cycle Cost (LCC). An integrated life cycle assessment (ILCA) determines the most suitable option. As a goal, a viable daylighting solution must offer good daylight performance, a lower Net Present Value (NPV) in its LCC and a better ILCA outcome than the existing setup.

The study combines field measurement to assess the base case, daylighting and lighting simulations for renovation scenarios, and calculations for LCA and LCC. First, the site was surveyed to retrieve geometry and to assess the existing electric lighting conditions. A 3D model was then created and verified against the field measurements of electric lighting. The verified model served as the base for three daylight renovation scenarios. Three different types of daylighting solutions were designed to provide similar daylight conditions indoors. Once the renovation scenarios were defined, they undergone an LCA and LCC. Lighting, LCA and LCC were eventually compared in an ILCA. Details are provided below.

The case study consists of an existing sports hall in Flyinge, Sweden (55° 45′ 17′ N, 13° 20′ 35′' E), part of a school serving students aged 6 to 12. Built in 1985 and largely unmodified since then, the 988 m² (43.6 m × 22.6 m), 7-m-tall building has recently undergone some updates consisting mainly of new flooring in blue wood (Figures 1 and 2). The sports hall currently relies solely on electric lighting. It has potential for the adoption of several daylighting solutions, thanks to its low-inclined flat roof and open surroundings.

Site visits were conducted to measure geometry, retrieve information on lighting systems, understand the building occupancy schedules, as well as to assess the current lighting situation.

The building is used daily from 08:00 to 16:00 for school activities. It is also used in the evenings (17:00 to 22:00) and some weekends by local sports associations for activities like football and badminton, for a total of 691 extra hours/year on top of the 08:00–16:00 occupancy. The current lighting situation is described in §2.1.2.

During site visits, the light environment for the existing building with the existing electric lighting system was surveyed and served to create a base case. The base case was assessed via the following.

1. Horizontal point-based assessments of illuminance with a calibrated Konica Minolta CL-70f spectrometer. Horizontal illuminances (Eh) were measured at 40 points across the floor (0.8 m height) (Figure 3).

2. Vertical measurements of illuminance (Ev) and spectral power distribution (SPD) were performed with the same instrument at selected points. The spectral information was used to measure the Correlated Color Temperature (CCT), the Color Rendering Index (CRI) and to calculate the Equivalent Melanopic Illuminance (EML) (Enezi et al., 2011; Lucas et al., 2014) provided by the electric lighting system. This information was later used for comparison with simulated values.

3. Optical properties of the surfaces, comparing visually the actual surfaces with those provided by the online database spectraldb.com (Jakubiec, 2022).

For the existing lighting system, technical specifications were retrieved from material provided by the buildings management and verified on site with inspections. The system consisted of fluorescent lighting for the large sports hall room, while new LED lighting was installed in the sitting area next to the hall. A schematic of the existing lighting system is provided in Figure 4. The large area with fluorescent lighting is indicated by the green measuring points in Figure 3, while the area with LED lighting is indicated by the purple points in the same picture. Since no daylight is provided, the energy use for lighting in the base case was calculated assuming the lighting was always on with occupancy, i.e. multiplying the power absorbed by the fixture for the occupancy time. This represents an underestimation of current lighting energy use, as no automatic control system is currently installed in the sports hall and, in real life, lighting may be occasionally left on even during unoccupied hours.

Following the field measurements, a 3D daylight model was created in Rhinoceros 7 and analyzed with Climate Studio v1 (Solemma LLC, 2022) and Alfa (Solemma LLC, 2025), aiming at characterizing both visual and non-visual aspects of lighting quality. The software was selected for several reasons. First, both programs rely on the Radiance simulation engine, one of the most validated daylight and lighting simulation engines (Mardaljevic, 1995; Reinhart and Herkel, 2000). Second, Climate Studio embeds natively a process for inserting TDDs, which were important to be tested for this specific case study, and which are optically complex to model otherwise. Third, for the non-visual aspects, Alfa provides direct integration into the Climate Studio workflow, without excessively sacrificing accuracy in comparison to similar software (Balakrishnan and Jakubiec, 2019; Pierson et al., 2022); the integration also allows working on a single model and workflow for visual and non-visual analysis, guaranteeing more consistency in the whole process. The model was first verified against the field measurements of vertical illuminance from electric lighting, showing variations below 2% between field measurements and simulated scenario. For example, for the view in Figure 2, the measured Ev was 338 lx, while the simulated one was 345 lx.

Considering that the building relies solely on electric lighting, the integration of daylighting was explored. The daylight design aimed to achieve homogeneous lighting (Uniformity Ratio, U₀ ≥ 0.6) and sufficient daylight illuminance, measured via daylight autonomy 300 lx (DA₃₀₀ ≥ 40%), in line with recommendations from EN 12464–1:2021 (CEN, 2021a). Note that the selected target value of DA₃₀₀ is lower than what is usually considered in office spaces (i.e. 50% across the space for LEED and WELL), to account for the custom occupancy schedule, which includes evening occupancy for local sports associations. The proposed daylighting solutions included:

1. Tubular Daylight Devices (TDD),

2. Skylight (SKL), and

3. sidelight windows (SLW).

These solutions have been chosen because they are proven to be suitable for increasing daylight and reducing energy use in the case of large, deep, spaces. Skylight, for example, has been successfully implemented in large commercial spaces (Yildirim et al., 2012; De Luca et al., 2018), while TDDs could improve daylighting in large spaces like stables (von Wachenfelt et al., 2015). Sidelight windows are also effective for lighting deeper spaces if they are placed high on walls. In addition, they are normally cheaper than skylights and TDDs, and they might result in less overheating than skylights.

The selected components are commercially available products, although their commercial names are omitted here, and values from Table 1 are retrieved from the official technical documentation of the products.

Once the base case model was verified against real measurements, the three proposed daylighting solutions for retrofit were implemented. The guiding principle was that the solution could be easily implemented in the actual building. This meant that the position of the system should not have interfered with the structural elements of the existing hall, the existing lighting layout or the existing ventilation system. As the second principle, as mentioned, the implementation was informed by two daylighting targets, which are coherent with the recommendations provided by EN 12464–1:2021 for sports hall (CEN, 2021a),

1. Daylight Autonomy, DA₃₀₀ ≥ 40% and

2. Uniformity Ratio, U₀ ≥ 0.6.

These two principles defined the number and position of systems for each tested solution. In the case of SLW, none of the solutions that were practically feasible could reach the target U₀ ≥ 0.6, see Table 4.

The results for daylighting simulations were also used to estimate the energy use for electric lighting for the simulated daylit scenarios. The energy use for lighting was calculated based on the Lighting Dependency (LD) defined as LD = 1 – DA₃₀₀, namely, each time that daylight falls below 300 lx, electric lighting is needed. The space was divided into three lighting zones, each controlled by an ideal daylight harvesting system with an occupancy sensor (5 min switch-off time). Using LD, operating hours and luminaire power, the electrical energy use for lighting was calculated. This simplified approach to the energy use for lighting calculation can provide quite robust results in this case, since automated daylight harvesting systems in real spaces with several occupants are more likely to function as designed (Pizarro and Gentile, 2019; Gentile et al., 2022). The standby losses of the newly introduced photosensors were not included, since they are negligible for larger systems and larger spaces (Gentile and Dubois, 2017).

Finally, the circadian potential of the space was also evaluated via Equivalent Melanopic Lux (EML). First, field spectral measurements provided the actual Spectral Power distributions (SPDs) for the existing fixtures. These SPDs were used in Alfa simulations. The simulations focused on a view, indicated with an orange cone in Figure 3. Such a view was selected as the worst-case scenario in terms of both daylight and electric lighting, being placed close to one of the side walls. An initial electric lighting simulation provided the EML value for the base case scenario (electric lighting). The geographical position was set to Malmö, Sweden, since it was the closest location with an available weather file. EML was computed for four key dates and three times (08:00, 12:00 and 16:00, except December, where 09:00, 12:00 and 15:00 were used). The dates and sky conditions were: 20th of March, Overcast sky (Vernal equinox); 21st of June, Clear sky (Summer solstice); 23rd of September, Clear sky (Autumnal equinox); 21st of December, Overcast sky (Winter solstice).

The introduction of the daylighting systems in the sports hall opens to environmental concerns due to changes in energy use for air conditioning and electric lighting during the use phase, as well as environmental impact associated with the production, installation and disposal of the daylighting components.

As mentioned, the electric lighting use was calculated with the LD approach. The heating energy was calculated using the degree hours method. The degree hours are the hourly temperature difference between the inside- and outside air, summarized in a year (Spinoni et al., 2014). Information regarding the supply temperature and the operating temperature of the building was given by the building manager. The total area of each daylighting product (TDD, SKL or SLW) was multiplied by the additional U-value increase and the degree hours of the application. The calculations considered the existing Ground Source Heat Pump (GSHP) with a Coefficient of Performance (COP) of 3. The additional heating costs due to the daylighting implementations were calculated and included in the Life Cycle Cost (LCC) and LCA using OpenLCA software to estimate impact over a 30-year lifespan. The lifespan matches that of the TDD and windows. Additionally, the building itself is unlikely to last beyond 30–50 years due to its age.

The three daylighting systems were compared using a cradle-to-gate LCA covering stages A1-A3 for all products. Operational energy use was also considered (B1, B6). The functional unit for the study was “the environmental impact of an illuminated sports hall floor.”

Since the TDD lacked an existing EPD, manual calculations were made based on EPDs for individual materials listed in the manufacturer’s product description. Quantities were derived from architectural drawings, product cut-sheets and detailed descriptions. Similar materials were used to estimate the TDD’s climate impact. The side windows and skylights had existing EPDs provided by the manufacturer, simplifying the environmental impact assessment.

Electricity and heating energy use were also added to the LCA, calculated using OpenLCA (GreenDelta GmbH, 2021) with the today’s Swedish energy mix, based on the European Reference Life Cycle Database (ELCD) (European Commission, 2024). Adopting a conservative approach, the energy mix was considered constant over the 30 years, although a decarbonization of the grid is expected in Europe. The excess energy use compared to the most efficient product was added to calculate the environmental impact of electric lighting and heating. The Centrum voor Milieuwetenschappen Leiden CML (Baseline) v4.4 method (CML, 2015) was used in OpenLCA, with no normalization or weighting applied.

The total environmental impact included these products, lighting and heating energy use. Once summarized, normalization and weighting were applied using a modified Shadow cost methodology from the Dutch government (CE Delft, 2010), based on the 2015 environmental prices (CE Delft, 2017) and excluding environmental indicators not in the EPD standard. Shadow cost, established by the Dutch government, is a system that assigns a cost to each kilogram of emitted environmental impact category equivalents through combined normalization and weighting. Often defined as “the highest acceptable cost society or the government is willing to spend on mitigating one unit of emissions” (de Klijn-Chevalerias and Javed, 2017), it is periodically updated to align with current Dutch political guidelines, with the latest edition dated 2023 (CE Delft, 2023). Its key advantage is translating environmental impacts into external costs, enabling internal and external project comparisons (van Harmelen et al., 2007). Table 2 lists the indicators and their normalized and weighted values.

The LCC assessment considered energy, installation and maintenance costs for each product’s anticipated 30-year lifespan. Full LCC parameters are detailed in Table 3 for comparison between products.

The electricity price was sourced from a local comparative site for fixed rates (Zmarta, 2022). With Sweden’s central bank prime rate at 0% and inflation around 4–6% at the time of this study, the real interest rate was estimated at 1% (Statistikamyndigheten, 2022). Electricity price growth was based on spot price data from 2015 to 2022 in Sweden. The annual growth was calculated by averaging each year’s monthly prices and comparing them year-over-year, yielding an approximate 8.3% annual increase. Adjusted for inflation, the growth was about 4%.

Preparatory and installation costs for the renovations were obtained from the national database Wikells Byggberäkningar AB (2022), with labor costs listed at 218 SEK per hour. Maintenance costs for cleaning windows and skylights were taken from a Swedish window cleaning company (Eriks Fönsterputs, 2022), assuming cleaning every two years. TDDs require no maintenance due to their dome design, according to the manufacturer.

Product prices were sourced from retailer websites or direct quotes. The cost of purchasing and installing a TDD was 15,000 SEK, provided by the manufacturer, with a total of 48 TDDs costing 720,000 SEK. The purchase of 28 skylights was 402,000 SEK, while the price for 44 windows was 227,040 SEK. Costs were sourced from retailers of the actual products.

Annual costs or savings involving electricity prices included the real interest and growth rates, applying a geometric gradient, as shown in the equation below.

where:

1. NPV = Net Present Value

2. A₁ = Annual payment, at year 1

3. g = Real growth rate for electricity prices

4. i = Real interest rate

5. N = Years

6. F = Future cost

If the real interest rate and price growth were equal, equation (2) was used. For future fixed costs, equation (3) (the compound equation) converted future costs to present values, such as maintenance. Inflation (k) was considered, reflecting the time value of money (The Swedish Energy Agency, 2017). Based on the financial situation at the time of this study (spring 2022), the LCC parameters for the first calculation were defined in Table 3.

Real interest rates (i) and price growth (g) were used, and parametric studies were conducted due to the unpredictability of future rates and prices. The real interest and growth rates for electricity ranged from 0% to 20%, with 1% increments. Electricity prices varied between 0.25 SEK and 5.25 SEK, in 0.25 SEK increments.

An integrated ILCA combined the LCA and LCC results to identify the most suitable daylighting option. Using normalized scores from the LCA and NPVs from the LCC, with initial weighting set at 90% LCA and 10% LCC, adjusted in 10% increments, the option with the lowest ILCA score was selected as the most appropriate. The adopted ILCA score equation was as follows:

The field measurements showed that the existing lighting system, including the newly installed LED fixtures, is over-dimensioned. The average values for the horizontal measurements resulted in Eh higher than 700 lx, well above the recommended 300 lx from EN 12464–1:2021. The areas mainly illuminated by fluorescent lighting provided an average horizontal illuminance of Eh = 675 lx and a CCT = 4050 K, with CRI = 73%, the small areas mainly illuminated by the new LED system provided Eh = 815 lx, CCT = 2984 K and CRI = 83%. Please note that CCT and CRI were measured on site and therefore they do not refer to the light source, but they are affected by reflecting surfaces.

The measured vertical light spectrum for the six defined view positions resulted in EML between 180 EML and 206 EML. This is just around the minimum recommended values from the WELL certification scheme, i.e. 150 EML.

Introducing daylight to match the target via DA₃₀₀ ≥ 40% and U₀ ≥ 0.6 required different numbers of daylighting components, see Table 4.

The results are also visually reported in terms of Daylight Factor (DF) in Figure 5. Figure 6 provides a rendering of the final TDD, SKL and SLW layout.

The resulting electric lighting use for the four cases, i.e. base case and three daylighting strategies, is provided in Table 5. It should be reminded that the values for the three implemented strategies correspond to an occupancy on-off system, 5-min delay, coupled to daylight harvesting with three photosensors for three lighting zones, using the same light fixtures as for the base case and excluding standby losses for the new sensors.

For the given scenarios, the light only needed to be on in full power mode for about 50–60% of the total time the space was occupied, including nighttime. During the remaining hours, it could be dimmed by roughly 25% during the “on” time.

The circadian potential simulations for the viewpoint, as in Figure 3, resulted in the EML and Ev provided in Table 6. Electric lighting alone (base case) provided 172 EML as simulated value, i.e. a slight underestimation with respect to the 193 EML measured on field. Since the model was validated with respect to the photopic illuminance, it is possible that the spectral reflectance of the simulated materials would slightly differ from that of the real one, resulting in the abovementioned underestimation.

Daylight provided between 5 EML with TDD on an overcast sky morning of December 21st and 1116 EML with SLW on September 21st with clear sky. On most occasions during clear sky dates, SKL and SLW provided constantly higher EML than the minimum 150 EML recommended by the WELL standard (IWBI, 2020).

The resulting energy use for heating for the three new scenarios is provided in Table 7.

The LCA results, which combine the environmental impacts of the product, energy use for electric lighting and additional heating, are summarized below. Figure 7 illustrates the environmental impact of the base case and the daylighting solutions. Over 30 years of operation, the base case exhibited the highest environmental impact of the four scenarios. The skylight had the lowest overall environmental impact, followed by the window and the TDD.

An overall comparison of the shadow costs from the LCA for the four cases is provided in Figure 8.

Shadow costs for global warming potential (GWP) were found constantly to be the highest for all tested solutions, including the baseline, followed by the acidification potential (AP). Other categories of environmental impact provided minor shadow costs. While the energy use for electric lighting provided the largest impact for the baseline and TDD cases, the LCA of SKL and WDW was largely impacted by the product itself, as well as the additional energy use for lighting and for heating with respect to the baseline.

For the base case, it is important to note that the energy use of electric lighting was the only influencing factor in this scenario, as no modifications or openings were made to the building fabric.

The LCA results for the TDD were affected by three factors: the product itself, the energy use for electric lighting and the additional heating required. Due to TDD’s low U-value, it required minimal additional heating. However, as the TDD provided the least daylight autonomy among the three options and it showed the highest environmental impact from electric lighting use compared to the other two solutions. Similarly, the skylight’s LCA results were influenced by the same three factors. However, the energy demand for heating played a significant role due to skylight’s higher U-value. For the window implementation, only two factors influenced the LCA results: the product’s environmental impact and the additional energy needed for heating. The window had the lowest energy use for electric lighting among the three implementations, which is why it was used as the baseline for the calculation, resulting in no excess lighting energy use. The main factor contributing to the environmental impact was the additional heating required, as the window had the highest U-value (1.16 W/m²K) among the options.

Table 8 presents the NPV for the adopted solutions assuming a reasonable real interest rate i = 1%, real growth rate of electricity price g = 4% starting from a price of 1.80 SEK/kWh and extending the life to n = 30 years.

The NPV depends on energy savings; therefore, the base case will never be profitable. In a mere cost perspective, and under the price and interest rate assumptions, the TDD results in the most profitable solution – arguably because of the absence of maintenance costs combined with higher energy savings for heating with respect to the other solutions –, followed by skylight and windows. Figure 9 provides a time perspective to the results. The TDD becomes profitable after 25 years, while the SKL and SLW become profitable after 26 and 28 years respectively.

Figure 9 refers to a specific financial situation. However, it is of interest to look at financial profitability under different scenarios. Increasing the real interest rate lowers the profitability of any renovation; with i > 2%, none of the daylighting solutions is profitable in financial terms (Figure 10a). However, an increase in the real growth rate of electricity makes any of the cases profitable in the set life span of 30 years when g > 3% (Figure 10b). Finally, having a higher electricity price increases the profitability of renovations due to energy savings. An electricity price of 1.50 SEK/kWh is sufficient to make the skylight solution profitable in 30 years, while the price should be at least 1.75 SEK/kWh for the window solution to reach profitability (Figure 10c).

The calculated ILCA results are presented in Figure 11. The weight given to LCC and LCA varied from 10% LCC and 90% LCA to 90% LCC and 10% LCA, with increments of 10%. Lower ILCA scores represent a better overall solution. The findings revealed that all proposed daylighting renovations would always be convenient from an ILCA perspective. The skylight was the most suitable option in eight of nine cases, when looking at ILCA. Only when the LCC was given 90% of the weight did the TDD become the most favorable choice, in line with the LCC results. The window implementation was overall more favorable than the TDD when the LCA was weighted at 60% or higher. In general, no daylighting renovations was outperforming the others in an ILCA perspective.

The results offer insights into both the existing building via on-site measurements and the viability of potential renovations. On-site measurements revealed high illuminance levels throughout the sports hall – more than double the recommended maintained horizontal illuminance – and promising potential for future daylight savings via lighting control systems, or even simple re-design of current electric lighting towards lower illuminances. Concerning daylighting solutions, the targeted daylight performance across all implementations was easily reached. However, the window installation did not meet the uniformity criteria, as its placement led to uneven light distribution primarily in the upper half of the hall. Daylight availability led to significant electrical energy savings of 44% for TDD, 49% for skylight and 50% for the windows. This is associated with a generally good daylight performance, with SKL outperforming the other solutions in both terms of overall illumination and uniformity.

The introduction of any of the daylighting solutions provided a boost to the circadian potential of the space. In particular, the SLW solutions provided the highest EML on average, throughout the year. Looking at median values, the SLW solution outperformed both TDD and SKL. This was mainly due to peaks during the autumn equinox, with EML over 1,000 EML. Since EML is calculated at eye level, such a high value may also suggest risk for glare. Glare analyses were not performed as current glare prediction models are empirically derived for office-based tasks, which are of a very different nature compared to the activity performed in a sports hall. Despite individual occupants using the sports hall irregularly, having a good circadian performance of the space can increase the quality of the space and the well-being of the occupants. Timing in circadian stimulation is of utmost importance and relatively short exposure to high EML at specific times of the day could have a larger impact on phase shift (Houser and Esposito, 2021). Thus, school children using the sports hall during mornings can have an extra benefit from the introduction of daylighting in the sports hall. In definitive, and in terms of daylighting quality, it could be argued that introducing daylighting solutions is always convenient and that the SKL solution represents the best among the three tested for this specific case study.

The LCA revealed that skylight had the lowest environmental impact, primarily due to effective daylighting that required only 28 units, significantly fewer than other implementations. This effective design led to a minimal environmental footprint over a 30-year period. Conversely, the window implementation faced higher heating demands, increasing its overall environmental impact. The LCA also showed that the environmental impact of materials is relatively low compared to long-term energy use. Normalizing and weighing different environmental indicators beyond just CO₂ equivalents allows for a broader understanding of impacts, such as those identified using the shadow cost method. The shadow costs indeed showed that while the GWP impact is the largest, the AP still represents a substantial part of the whole environmental impact.

The limitations of the adopted LCA approach merit discussion. First, the cradle-to-gate approach conceals the environmental performance of the solutions if re-use is planned. Glazing assemblies might be more easily reused or recycled than other components in other daylighting systems. Second, the conclusions are based on the Swedish energy mix, with a prevalence of green electricity. If other energy mixes or different heating systems are considered, the results could be substantially different. Yet the limitations are a further demonstration that an overarching and critical approach to renovation is always needed before jumping into decisions.

The LCC results highlighted substantial potential savings using daylight, despite the long payback periods, in the order of 28 years. A key recommendation from this study is to consider daylight integration early in the design process to maximize economic benefits. The extended payback periods were largely due to the inclusion of costs for dismantling the existing envelope in the installation expenses. Minimizing or avoiding such dismantlement could significantly reduce the payback period, enhancing the financial feasibility of daylight solutions. Besides costs, the assumptions on NPV were found to be of particular importance to define the financial viability of the investment. While a real interest rate of 2% – which is not very high – would not make any of the daylighting strategies economically remunerative, a slight increase in electricity price would lead to opposite scenarios. While discussing solutions, it is important to provide different potential economic scenarios before making any decision and weighing them together with the positive quality impact of introducing daylight in such spaces.

Taken alone: (1) the sidelit windows would perform best in terms of daylight and circadian potential, despite lack of uniformity; (2) the skylight was found to be the most suitable from a LCA perspective; (3) the TDD was the most cost-effective option from a life cycle cost perspective. Therefore, no solution outperforms the others for all given criteria. In this perspective and in line with what was shown by other studies (Schneider-Marin et al., 2022; Bianchi et al., 2021), it was interesting to propose an ILCA which allows for discussing the scenarios, putting emphasis on the different criteria. The ILCA provides food for discussion in a hypothetical decisional process, in this case showing that some solutions, like the TDD, were suitable only when the emphasis was on the financial aspect, while the suitability of other solutions depended on the weight given to the LCA or the LCC. In general, renovating with any of the daylighting solutions was better than keeping the building as it is, for all considered criteria.

In this study, an ILCA approach was proposed and applied to a daylight renovation, using a sports hall as an example case study. Three different scenarios hypothesizing renovation with new daylighting systems were created. The scenarios – TDD, skylight and side windows – were analyzed in terms of daylighting quality, circadian potential, environmental impact and financial viability.

For the specific case study, it was found that:

1. For visual quality, all three tested daylighting strategies – Tubular Daylight Device (TDD), Skylight (SKL) and Window (SLW) – achieved the target daylight autonomy (DA₃₀₀ ≥ 40%), though SLW did not meet the required uniformity (U₀ ≥ 0.6). Annual electric lighting energy savings reached 44% (TDD), 49% (SKL) and 50% (SLW), despite SLW’s uneven distribution. Circadian potential improved significantly, with SLW yielding median EML levels above 150 EML and peaks exceeding 1,000 EML during equinox mornings, suggesting a strong biological impact, especially for morning-active school children.

2. For environmental impact, the skylight had the lowest 30-year environmental impact, needing only 28 units and benefiting from high daylight contribution. In contrast, the SLW required more heating due to its U-value (1.16 W/m²K).

3. Economically, TDD had the shortest payback (25 years), while SLW and SKL became profitable at 28 and 26 years, respectively, under low interest and rising electricity costs. The Integrated ILCA identified SKL as the most balanced solution.

Therefore, under the given assumptions for the LCA and LCC analyses – including the LCA phases, national energy mix, interest rate and electricity price – none of the daylighting solutions clearly outperformed the others across all criteria. This reflects a fundamental trade-off: daylight quality, environmental performance and financial feasibility are not automatically aligned, at least in this case study. The ILCA analysis was thus instrumental in illustrating how different solutions serve different priorities. By integrating multiple performance aspects, ILCA can better support decision-making and help building owners reach informed and economically sound outcomes.

Although a simple weighting method was applied here to compare alternatives, the findings highlight the need for ILCA-based approaches in daylight renovation projects – approaches that jointly address all three pillars of sustainability: environmental, social (via light quality and health) and economic.

This is particularly important because daylight interacts with buildings in non-trivial ways: higher daylight availability may improve light quality only up to a point – beyond which glare becomes an issue – while also impacting thermal loads, energy use and occupant comfort. Optimizing across these interconnected aspects is essential, and such integration should begin already in early design phases if truly sustainable buildings are to be achieved. Optimizing for all aspects together is therefore imperative.

This study presents several limitations. First, the study is limited to an example case study, and it includes a limited set of renovation packages; future work may look at more extensive optimization processes using a larger dataset and optimization algorithms. Optimizations and weights might be tuned according to the decision-maker's requirements. Second, although the LCA was performed following best practices, LCAs tend to be inherently inaccurate due to a lack of consistency in EPDs. The same consideration should be done for the costs associated with different environmental impacts in the shadow cost methodology. Finally, general limitations related to the process of modeling and assessing building performance (the “performance gaps”) should always be considered.

Carl Laursen: Conceptualization, methodology, software, formal analysis, investigation, data curation, writing – original draft, writing – review and editing and visualization. Niko Gentile: Conceptualization, methodology, resources, writing – original draft, writing – review and editing, supervision, project administration and funding acquisition.

This study did not involve living subjects, human or animal. All data were gathered through technical environmental assessments and computer simulations. Therefore, no ethical approval or informed consent was required, in accordance with national guidelines. Grammarly and ChatGPT were used to copy-edit the Introduction and Conclusions to improve the language and readability.