**2. Methodology**

In this section, the structural system components and configurations were discussed, followed by a description of the 3D printing technology utilized to construct the house understudy. Moreover, the standard methods of the environmental and financial life cycle analyses were presented.

#### *2.1. Structural Systems*

A single-storey detached house located in the UAE was selected as a case study. Figure 1 shows the plan and elevation layouts of the selected house with a net floor area of 90 m<sup>2</sup> and total height of 4.5 m. The proposed structural systems include (1) conventional construction method using cast in place concrete walls and flat slab with beams and columns, and (2) additive manufacturing using self-reinforced printable mortar. It should be noted that the construction time frame of the 3D printed house was approximately 2 weeks, whereas the conventionally built house was 4 months based on local engineering contractors. The timeframe excludes the HVAC, plumbing, and finishes works as they are similar in both houses.

**Figure 1.** The technical drawings for (**a**) ground floor, (**b**) Site plan, (**c**) section A, and (**d**) section B.

Table 2 shows the details of the structural elements utilized for conventional concrete construction. The columns and beams have a cross-sectional area of 800 and 1600 cm2, respectively, whereas the slab has a total area of 376 m2. Wood formwork was utilized in construction of the columns, beams, and slabs of 3.8 m2, 47 m2, and 400 m2, respectively. There are 0.03, 0.04, and 0.245 m<sup>3</sup> of columns, beams, and slabs per m2. The design of the steel reinforcement, confinement steel, and stirrups were conducted according to American Concrete Institute (ACI) standards [19]. Moreover, the considered primary loads in this study were the typical dead and live loads defined by American Society of Civil Engineers (ASCE) 7–10 [20].


**Table 2.** Dimensions and reinforcement of structural elements.

\* The design details include main and secondary reinforcing rebars.

The specifications and properties of the cementitious mortar used for conventional concrete and 3D printing mixtures are summarized in Table 3. The conventional concrete mix has cement, sand, and aggregates ratio of 1 to 1.5 to 1.3, respectively, while the cementitious 3D printing mortar consists of 70% sand and 30% binder (cement and additives) [21]. Moreover, the mix of the 3D printing mortar is characterized by low sulphate and chloride content which was designed for structural and non-structural elements.

**Table 3.** Properties of 3D printing and conventional construction materials \*.


\* Compiled from [21] and \*\* [22].

#### *2.2. Additive Manufacturing Technology*

The application of a large-scale 3D printed structure entails using an extrusion method, in which the structure was built by adding layers of the prepared mortar through a nozzle. The digital STL (STsereo Lithography) formatted file was converted into several 2D layers by means of CyBe CHYSEL software [21]. Moreover, Table 4 summarizes the input parameters required for the operation of the mobile 3D printer. Furthermore, the printing process was regulated through a control unit which operates the mixing system to pump the mortar through a hose into the robotic arm. The mortar was added layer by layer at the specified coordinates via a 40 mm nozzle. The 3D printing filaments were characterized by a zigzag pattern and the printed walls were hollow (39 cm).



#### *2.3. Life Cycle Analysis*

The environmental impacts and burdens on the ecosystem of production, construction, operation, and disposal stages over the life cycle of a system was quantified using the LCA systematic framework. The international organization for standardization (ISO) developed ISO 14044 and ISO14045 to unify the approach of evaluating the load on the environment, address the resulting ecological impacts and identify potential performance enhancement over the lifecycle of the systems [22,23]. Two LCA approaches are commonly investigated in the construction industry, namely, cradle to grave and cradle to site. The first method includes all materials and processes in a comprehensive assessment, while the second approach focuses on certain aspects of the construction project such as the materials [17]. In this study, a cradle to site approach was selected and the LCA was performed in four stages including, goal and scope, life cycle inventory (LCI), and life cycle impact assessment (LCIA) analysis, and results interpretation. Stage one of the LCA involves defining goal and scope as well as the system boundaries and functional unit. The LCI phase includes collection of data, while the third sage (LCIA) examines the contribution of these data to selected impact categories. Stage 4 involves assessment of the results and identifying study limitations. SimaPro 9.0 developed by PRé Sustainability was utilized to implement the LCA framework using Ecoinvent 3.0 [24].

#### 2.3.1. Goal and Scope Definition

The goal of this study is to evaluate the environmental performance of a 3D printed house compared to conventional construction techniques. Measuring the functionality of both construction techniques output was achieved by selecting a reference or a functional unit; 1 m<sup>2</sup> of the single-storey house surface area was selected for simplification of inventory data calculations. Figure 2 shows the boundaries of the examined systems including, production and manufacturing of materials, construction, operation, maintenance, and end of life phase. However, the LCA assessment was limited to material extraction, construction, energy consumption, and transportation during the operation phase. Similar components in both structural systems were excluded i.e., earthworks, HVAC systems and finishes. The labour and end of life phase were excluded from the study as they were found negligible [17]. Moreover, all of the reviewed literature (Table 1) excluded the end of life or demolition phase as a result of lack of available data.

**Figure 2.** System boundaries of 3D printing and conventional construction of the examined house.

#### 2.3.2. Life Cycle Inventory

The input data related to 3D printing and conventional construction were gathered from local suppliers, Ecoinvent database and the literature. Such technical data include foreground components such as quantity of materials, transportation, and energy consumption. Moreover, background data of the environmental burdens were assigned to the foreground processes and components. Table 5 lists the inventory data of the examined structural systems, in which energy consumption of the equipment utilized on-site can be measured from the power demand and operation time of such machinery.

**Table 5.** Life cycle inventory data of the examined systems per functional unit.


\* [25] \*\* [2,26] \*\*\* [27]..

#### 2.3.3. Energy Consumption

The energy consumption rates in the construction sector reach up to 40% of the total energy demand [28]. The primary electricity consuming sources are the cooling systems as a result of the harsh climate of the UAE with temperatures reaching up to 48 ◦C, hence the construction sector is constantly exploring efficient heat insulating materials to prevent overheating and humidity increase. The European commission has reported that buildings are responsible for at least 40% of the total energy consumption Particularly, air conditioning is a major energy consuming element in a building, hence reduction of cooling load demand by thermal insulation through construction materials inducing low heat transfer can save up to 50% of the building energy demand [29]. The energy savings for the 3D-printed and conventional concrete house were calculated based on the

difference between the microclimate and the air temperature surrounding the structure as well as the thickness of the structural elements (external walls and roof). The ISO standard (EN ISO 6946:2008) reported the key factor to indicate the thermal properties of the building is heat transfer (U) in which lower U-value indicates higher energy savings [30]. The U-value [31] and the energy transfer or heat flow (Q) [32] were calculated using Equations (1) and (2) [33,34]:

$$\mathbf{U} = \frac{\mathbf{k} \times \mathbf{A}}{\mathbf{l}} \tag{1}$$

$$\mathbf{Q} = \Delta \mathbf{T} \times \mathbf{U} \times \mathbf{A} \tag{2}$$

where U is the thermal transmittance (W/m<sup>2</sup>·K), k is the thermal conductivity of a material (W/m·K), A is the plane area of (m2), l is the thickness of material (m), Q is heat flow (W), and ΔT is the temperature difference between external and internal structural element surface (◦C). The heat transfer through individual rooms of the house, the windows, and doors was calculated. The design temperature outside and inside the house was specified by local guidelines as 46 ◦C and 24 ◦C, respectively. Moreover, the U-value of the floor and roof slabs were obtained from local standards and they were compared to ASHRAE (American society of heating, refrigerating and air-conditioning engineers) specifications based on perimeter to area ratio and thermal resistance values [33,34].

#### 2.3.4. Life Cycle Impact Assessment

The environmental impacts of the digitally fabricated and conventionally built house were estimated using ReCiPe 2016 V1.03 midpoint (H) indicators [35]. The method represents the impacts of a global representative and addresses 18 different categories. The impact mechanisms include climate change or global warming potential (kg CO2 eq) ozone layer depletion (kg CFC-11), terrestrial acidification potential (kg SO2), marine eutrophication (kg N), freshwater eutrophication (kg P), human toxicity (kg 1,4dichlorobenzeen), particulate matter formation (kg PM2.5), ionizing radiation (kBq Cobalt-60), photochemical oxidant formation (kg NMVOC), terrestrial, freshwater, and marine ecotoxicity (kg 1,4dichlorobenzeen), agricultural and urban land occupation (m2), freshwater depletion (m<sup>3</sup> water consumed), mineral resource depletion (kg Copper (Cu)), and fossil fuel scarcity (kg oil) [35]. The impact categories represent the effect on the environment and are based on weighted and normalised factors [36].

#### *2.4. Life Cycle Costing Analysis*

The financial viability of 3D printing and conventional construction techniques was investigated by calculating the construction and energy use costs. The capital cost of the examined projects included procurement and manufacturing of construction materials e.g., cement, steel, wood, aggregates, and admixtures, as well as construction activities. The present value (PV) of the electricity costs of the systems was estimated for a period of 50 years, which was carried out via LCCA framework to estimate the present worth of the energy consumed in the 3D printed and conventionally constructed house. Moreover, the time value of the cashflow was considered in this study using a local-based discount rate of 3% [37]. Equation (3) is used to calculate the present value [38]:

$$\text{PV} = \sum\_{\mathbf{t}=1}^{\text{T}} \mathbb{C}\_{\mathbf{o},\mathbf{t}} (\mathbf{1} + \mathbf{r})^{-\text{t}} \tag{3}$$

where Co is the cash outflow (USD) of year t, r is the discount rate (%), and T is the lifespan of the project.

#### *2.5. Eco-Efficiency Analysis*

Selection of an optimum alternative and identification system trade-offs can be accomplished through an eco-efficiency analysis. Such analytical framework functions by agglomerating LCC and LCCA results, which are plotted into a single portfolio [23]. The

ratio method is the most commonly used approach to determine the eco-efficiency of a system or a product [39–41]. In this study, the ratio method was employed which is defined as the ratio of economic indicator to environmental performance of the examined system as shown in Equation (4) [41].

$$\text{Eco} - \text{efficiency} = \frac{\text{Enviornmental Performance}}{\text{Economic Value}} \tag{4}$$

The Environmental indicator in this research study was retrieved from the LCA SimaPro software represented by a normalized and weighted single value aggregating all the midpoint categories. Moreover, the present value was utilized which corresponds to the economic indicator of each assessed system. An eco-efficiency portfolio combining environmental and economic scores was plotted for the selection of the most eco-efficient system and assessing the trade-off among the studied alternatives.

#### **3. Results and Discussion**

#### *3.1. Environmental Analysis*

The LCA results analysed in this section represent a comparison of additive manufacturing and conventional construction techniques in terms of the environmental impacts. The environmental impacts of the studied scenarios were calculated via SimaPro in 4 stages—characterization, damage assessment, normalization, and weighing [24]. During the first stage (characterization), the materials were multiplied by a factor that represents the relative contribution. The damage assessment facilitates the use of endpoint categories, where impacts with the same units can be added. Normalization stage enables comparison among scenarios in which the impacts are divided by a reference. The weighing phase is typically performed by multiplying the impact categories with a factor and adding them to result in a single score. This score is an indication of the total impacts. Table 6 provides detailed environmental performance scores for each impact category of the 3D-printed and concrete-based house. Most impact categories had significantly higher values for the conventional construction method. Among the highest scored impacts in the conventionally built house were global warming, non-carcinogenic toxicity, water consumption, carcinogenic toxicity, and fossil resource scarcity. Cement production contribution to global warming potential (1154.2 kg CO2 eq) was approximated to be 70%. Moreover, reinforcing steel production and manufacturing comprised 98 and 97% of the total emissions of non-carcinogenic and carcinogenic toxicity with relative impact of 675 and 169 kg 1,4-DCB, respectively. Furthermore, fossil scarcity (150 kg oil eq) was attributed to the manufacturing of steel (60%) and cement (38%), and the high-water consumption was mainly due to addition of water during concrete manufacturing. The Global warming potential and water consumption had relatively high impacts for the 3D-printed house. As for the concrete constructed house, global warming potential (609 kg CO2 eq) was high due to production and manufacturing contributing 97% and water consumption with a volume of 184 m<sup>3</sup> per functional unit was attributed to water demand during 3D mortar preparation. The endpoint indicators were represented by a single score that combines all the inventory results in one factor. For the 3D-printed and the conventional house, the human health category had substantially higher impacts compared to effect on ecosystem and natural resources indicators. Human health category caused 93 and 88% of overall emissions of the conventional construction and 3D printing scenarios, respectively.

The obtained results from SimaPro were normalized and weighted to provide holistic assessment. Normalization enables for a coherent interpretation of the characterized environmental impact categories through referring to a reference scheme, followed by weighting which emphasizes the relative significance of the impact indicators. Figure 3 shows the relative environmental impacts of the examined systems analysed based on different impact categories. It is evident that 3D printing has an overall lower impact across all categories. The 3D printing scenario performed more than 50% better for the majority of the categories which may be attributed to the material efficiency compared to the

conventional scenario. Typically, conventional building requires formworks and reinforcing steel, which are absent in the 3D printing scenario. Therefore, all emissions related to the production, manufacturing, transportation, and fabrication of materials are reduced. The damage to the ecosystem was minimal where the midpoint categories pertaining to freshwater marine, and terrestrial species had relatively low percentage (0–7%). Though all categories of 3D printing had lower impacts, the water consumption category was only 20% better for the 3D printed house due to high water use during cement production processes and electricity generation, which is common to both construction methods.

**Table 6.** Environmental inventory results of the examined structural systems.


**Figure 3.** Relative environmental impacts of 3D printed and conventional constructed houses.

> In the digitally fabricated house, cement production phase contributed (more than 95%) to most of the impact categories i.e., global warming, ozone depletion, terrestrial acidification and ecotoxicity, human carcinogenic impacts, and fossil and mineral resource scarcity as shown in Figure 4. Moreover, material extraction and production of the utilized admixtures was a major contributing process to land occupation, freshwater eutrophication,

ionizing radiation, marine and freshwater ecotoxicity, and non-cariogenic human effects, with 99, 98, 97, 61, and 40%, respectively. Electricity and transportation obtained the lowest ratio in all environmental impact categories with impacts ranging between 0 to 2%.

**Figure 4.** Contribution of 3D printing processes to the overall environmental impact.

The contribution of the different impacts i.e., production of cement and steel, manufacturing of concrete, transportation, as well as electricity production are shown in Figure 5. The cement production shows the highest contribution in all impact categories due to significant consumption of raw materials and energy, the greenhouse gas emissions during manufacturing phase, and the release of bulk amounts of waste. Moreover, the environmental analysis revealed that reinforcing steel production and manufacturing processes had a primary impact on freshwater eutrophication (99%), land occupation (98%), terrestrial and marine ecotoxicity (93%), carcinogenic, non-carcinogenic and freshwater ecotoxicity (89%), fossil resource scarcity (60%), and global warming (41%). Similar to the conventional house results, the electricity scored the lowest in all categories except ionizing radiation (11%). Overall, the exploitation of materials, energy use, and transportation during manufacturing of concrete components poses the highest environmental risks as can be deduced from Figure 5.

**Figure 5.** Relative contribution of conventionally constructed house processes to the environmental impact.

The results of this study agree with the outcomes of [2,12,17], which reveals that 3D printing structures outperform the conventional construction methods in terms of overall environmental impacts. The main difference in this study was conducting the analysis for the entire house, whereas [2,12] studied the impacts on individual elements (wall, roof, and a concrete slab) with varying design complexities and included the operation phase for the self-shading wall element. Moreover, the 3D printing mortar ratios and components in this study was tested for an implemented project in the UAE, while Agustí-Juan and Habert [2] adopted a fiber reinforced concrete from the literature and Alhumayani et al. 16 tested out three different mixes also compiled from the literature and compared the results. Furthermore, Agustí-Juan et al. [12] designed a high performance 3D printing concrete which was found to increase the GHG emissions when compared to conventional concrete mix.

#### *3.2. Operational Energy*

The cooling energy demand for the 3D-printed and conventionally constructed house was calculated considering the thermal transmittance of the construction mortars. Table 7 summarizes the cooling systems calculation results for the 3D-printed and conventionally constructed house. Overall, the total heat transfer (gain) of the conventional building system was 5% more than the 3D printed house. The 3D printed house acquired less heat gain due to higher material thickness and thermal transmittance (K). In other words, the lower thermal conductivity and thickness of materials the lower heat transmission. Another contributor to low heat conduction is U-value, where the slabs of a 3D-printed house had lower U-values compared to the conventional concrete house. On the other hand, the insulating properties of the 3D-printed wall including an air cavity had a much higher U-value (3.75 W/m<sup>2</sup>·K) which is in close proximity to the concrete wall (3.6 W/m<sup>2</sup>·K).



\* The wall U-value includes air cavity with thickness 0.04 m and R of 0.12. \*\* The total heat gain includes heat from doors and windows.

#### *3.3. Economic Assessment*

The economic analysis findings of the selected structural systems are summarized in Table 8. The results comprise capital costs of materials (local-based) including civil works and operational expenditures of cooling systems. The conducted present value over a 50-year design period indicates that conventional construction technique was the most expensive alternative (USD81,064) which was double the cost of the 3D printing. This can be attributed to the cost of concrete, and formworks which comprise 51 and 24%, respectively. The capital expenditures of concrete are associated with the purchase and manufacturing of various sub-components, mainly aggregates (USD10,795). Although the steel cost rate (USD500/ton) was the highest, it had the least contribution to the overall cost. On the other hand, the 3D printing technology was found to be 49% cheaper than the conventional construction scenario. The 3D printing excludes multiple aspects including construction components, e.g., concrete and formworks, as well as labor cost, thus reducing the overall capital costs. These results are in line with [18], where the 3D printing of houses contributed to 35% savings compared to conventional construction.


**Table 8.** Capital costs of construction components and operational expenses.

Note: Positive present values signify revenues, whereas negative values represent costs.

#### **4. Eco-Efficiency Analysis**

The depicted results of economic and environmental performance ratios were plotted in an eco-efficiency portfolio as illustrated in Figure 6. The top-right corner distinguishes the low eco-efficiency alternative, while the bottom left corner of the plot area identifies the high eco-efficiency option. The conventional construction house had significantly lower eco-efficiency compared to 3D-printing. Upon comparing the operation phases of both houses, the results reveal similar eco-efficiency scores, which coincides with the LCC and LCA analyses. Moreover, the eco-efficiency index diagram orders the alternatives from the highest (bottom) to lowest (top) eco-efficiency. The 3D printing method was found to be the highest and conventional construction acquired the lowest eco-efficiency. The findings of eco-efficiency analysis showed that operation phase alone was negligible in the selection process of the optimum alternative, nevertheless the combined construction and operation phase revealed 3D-printing as the most eco-efficient option.

**Figure 6.** Eco-efficiency portfolio of 3D-printed and concrete-based house construction and operation phases.

#### **5. Sensitivity Analysis**

Several factors such as system boundaries, assumptions, and accuracy of inventory data affect the certainty of LCA and LCC results. Moreover, the 3D printing technology is still in the exploration and development stage and the data were compiled from the literature. A sensitivity analysis was conducted to account for the uncertainties in this study where the selected parameters are listed in Table 9. Different 3D printing binder mixtures were evaluated in the analysis to investigate the environmental impact of cement and coarse aggregates as they acquired the highest scores in the LCA results. The conventional concrete mix was also evaluated to investigate the effect of varying concrete and steel quantities [2,42].


**Table 9.** Parameters utilized in the sensitivity analysis for 3DCP and Conventional scenarios.

> \* Adapted from [15] \*\* [2], and \*\*\* [42], \*\*\*\* The energy consumed by machinery.

The concrete, steel, and cement production accounted for the highest environmental scores in the performed LCA. Figure 7 illustrates the results of the sensitivity analyses for the different 3DCP and Conventional mixtures. The results are presented relative to the conventional base scenario which obtained the highest impacts in all categories. The analysed mixtures had relatively small impacts contributing to 0–3% in all categories. Nevertheless, the 3DCP mix 1 and 2 contributed to the highest water consumption (474 and 391 m3, respectively), followed by conventional mix 1 (390 m3), conventional base scenario (233 m3), the 3DCP base scenario (184 m3), and the least water consumption was attained by conventional mix 2 (110 m3). These results led to the conclusion that reducing cement quantities in 3DCP binder can reduce the overall environmental impacts by 90%. In conventional construction techniques replacing some concrete elements with bricks (such as conventional mix 2) can also reduce the environmental deterioration.

**Figure 7.** Sensitivity analysis results of different conventional and 3D concrete printing (3DCP) mixtures.

The LCC results of the different mixtures reveal significant differences from the original scenarios (Table 10). The 3DCP mix 1 and 2 showed almost similar results with a decrease of 20% from the original mix. This decrease can be attributed to the reduction of cement in mix 1 and mix 2. Conventional concrete mixtures 1 and 2 obtained a total cost of USD 33,073 and 31,451, respectively which is almost 60% less than the base scenario. Moreover, the cost of the 3D printer was added to the 3D printed house scenario while keeping all the other parameters constant. The present value was found to be USD 225,391 (82% increase in expenditures). Since the technology is still in the exploration stage, a renting cost is ye<sup>t</sup> to be accounted for in future 3D construction projects. Different electricity tariffs ranging between 0.07 to 0.1 were investigated. For low electricity tariffs, the costs of the 3D printing scenario decreased by 5% and increased up to 25% for higher ranges. Similarly, the costs of the conventional scenario decreased by 7% and increased up to 7% for higher ranges.


**Table 10.** Life Cycle Costing of the different sensitivity analysis alternatives.


Data uncertainty and limited availability typically affects the life cycle assessment results. Figure 8 shows a +10% variation of the LCC and LCA parameters studied in the current research. The figure revealed a correlation of operation of both 3D printed and conventional scenarios. Nevertheless, the construction of conventional system had the greatest environmental impact and greatest cost with the variation.

**Figure 8.** Uncertainty analysis of with +10% variation of 3D printing and conventional construction scenarios.

#### **6. Study Limitations**

Based on the conducted structural, environmental, and economic assessments, 3D printing is a viable alternative to conventional construction techniques. However, the findings of this comparative study were limited due to the unavailability of some important data, such as, (1) characteristics of the mortar used in 3D printing process, (2) varying ratios of conventional concrete ingredients, (3) limited number of investigated structural elements, (4) exclusion of sub-structure system and end of life phase, and (5) the common processes and components among the examined alternatives were not included, thus only relative environmental impacts were quantified, (6) inadequacy in 3D printing speicifc processing and (7) data inventory was calculated from diverse sources as a result of lack of data.
