The Environmental Sustainability Study of an Airport Building System Based on an Integrated LCA-Embodied Energy (Emergy)-ANN Analysis

Abstract

From a global perspective, the ecological sustainability of building systems has always been a hot research topic, especially in China, where the annual amount of new construction is nearly half of the world. The difficulty is making a complete and accurate ecological assessment of the building system. This study has designed and adopted the LCA-Emergy-ANN framework to assess and analyze an airport building system for sustainability. The results demonstrate that building material emergy and operational stage emergy play a critical role and account for 92.4% of the entire emergy, which are the primary contributors. As the vital indicator, the emergy sustainability index (ESI) is 0.669, which is unsustainable (The eligibility standard is 1). Simultaneously, to ensure the accuracy of the data results, sensitivity analysis was performed. The artificial neural network (ANN) was used by integrating the LCA method and emergy approach to predict the sustainability trend in the long run. In the end, the optimization strategy is proposed to enhance the sustainability of the building system.

1. Introduction

As an important part of the global ecosystem, building systems are indispensable for human habitation and coping with climate deterioration. With the growth of population and environmental degradation, the concept of ecological architecture is proposed and promoted [1]. For ecological building systems, it means that ecological points and methods are applied to the field of architecture to improve the ecological design and evaluation to achieve the intersection of architectural systems and ecological approaches [2,3].

For the investigation and exploration of ecologically sustainable buildings, many scholars have made in-depth studies, which mainly focus on five aspects, including ecological indexes [4,5], assessment views [6,7,8], sustainability perspective [9,10,11,12], building renewal viewpoint [13,14], intersecting methodology [15,16,17], etc.

The type of research involves many kinds, such as green buildings [4,5], built assets [6], residential buildings [7,8,9], building design process [10], building review (new buildings\in-use buildings\single building\group of building\building element) [11], school building [12], old building [13], intelligent building [14], mixed building type [15], green building example [16], high-rise timber building [17], etc.

Studies have explored the relationship between resilience indexes and building systems [4]. Depending on three-class element network indicators, the building systems were implemented and displayed for the ecological hierarchy [5].

From an evaluation point of view, it is a hot spot for building systems. Based on the decision-making framework, the building system was assessed for sustainable development [6]. A new rapid evaluation approach was utilized to obtain the building system's assessment result, showing further possibility [7]. With the complex construction process, a social-based evaluation methodology was selected for better managing sustainable results [8].

The sustainability perspective is a key point in the ecological building field. In response to local climate changes (tropical climate), benchmark sustainability standards are emphasized [9]. The LCA approach was used for building systems through environmental, economic, social, and other comprehensive considerations [10]. Several researchers have taken a closer look by reviewing the literature on sustainability investigation in the building system [11,12].

The research of architectural sustainability renewal is also the focus of scholars [13,14], which connects the present and future building development trends.

In addition, various methods are used in the field of ecological study of building systems, such as the bibliometric mapping approach [15], green system view [16], reassessment method [17], etc.

Therein, as an ecological analysis, the emergy method has been adopted and discussed in many fields, such as the ecological field [18], economic aspect [19], agricultural assessment [20], industrial study [21], urban view [22], water pollution treatment [23], etc. The cross-study between the architecture field and the emergy method is relatively new and has been rising gradually in the last decade. As a new method, LCA-Emergy was provided for the sustainability assessment in the building system [24], including building material and construction process, etc. [25,26,27,28]. In addition to common building types, large building systems and zero-energy building systems are also considered for ecological characteristics [29,30,31]. Some scholars have made various explorations involving renovation effect [32], building uncertainty [33], and cross approach [34].

In addition, the whole life cycle approach is widely used in architecture. Several authors tried to integrate LCA and emergy methods to analyze the sustainability of the building system. By reviewing the literature of the past five years, only a few articles cover the LCA-Emergy method. For example, for building refurbishment, several strategies have been conducted based on the emergy-LCA method [35]. A residential building was selected for sustainability investigation in light of emergy analysis [36]. As a necessary part of the building system, the building cement material system has been concerned and analyzed to use an emergy view [37]. As a particular form of architecture, highway engineering has also been surveyed by emergy evaluation [38].

In addition, some defects restrict the development of this direction, including an old emergy benchmark, inadequate evaluation index, an incomplete emergy assessment calculation framework, etc.

With the development of artificial intelligence computing, artificial neural network (ANN), as a new method, is gradually being applied to many domains, which are also increasing applications in building systems [39,40]. Until now, through a literature review and verification, there has been no LCA-Emergy-ANN investigation and exploration of building systems, which is just the research direction in this study. This paper aims to figure out a series of questions, including:

• How is the LCA-Emergy-ANN method framework designed?
• How is the emergy boundary of the whole life cycle of the building system controlled?
• How to define and select evaluation indicators to assess sustainability?
• How to analyze the sensitivity of emergy analysis results?
• What are sustainable building system measures? How to determine the effect of improvement?

Based on the above five questions, this paper conducts the corresponding study and discussion, which provides references for architects and engineers.

2. Methodology and Case

Section 2 has four subsections: analysis framework, emergy introduction, artificial neural network, and case study.

2.1. Analysis Framework

Figure 1 demonstrates the research framework, mainly consisting of two parts, which are the LCA-emergy method and the ANN-emergy approach, respectively.

Based on LCA-Emergy analysis, seven types of inputs have been adopted to explore the sustainability of building systems, including solar, material, electricity, water, diesel fuel, gasoline, and human labor. Given the ANN-Emergy approach, the emergy research method of the artificial neural network was leveraged to predict the sustainable development of building systems in the long run.

2.2. Emergy Introduction

The Emergy concept has been defined as embodied energy, which differs from the energy view. H.T. Odum first put forward to evaluate the sustainability of ecological systems [41], which has an obvious advantage in building a platform for environmental assessment. As available energy, it realizes the alignment of various input items based on one unit standard to compare. The unit of emergy is the solar joule (sej), which can be transformed from any kind of energy. Among them, unit emergy value is the core point to measure and judge the hierarchy in the building system, which has three forms, including emergy per unit of energy (J), substance (g), and economic ($). Hence, the unit of UEV is sej/j, sej/g, and sej/$.

2.2.1. Basic Emergy Calculation

Emergy analysis can basically be divided into five main steps: (1) Basic data acquisition and collection; (2) Emergy system diagram design; (3) Emergy analysis table calculation; (4) Comprehensive emergy indicators analysis; (5) Final evaluation and strategic analysis in the building system.

The emergy T can be calculated by Equation (1).

$$T={\displaystyle \sum _{j=1}^n{E}_j\times }{T}_{r_j}$$

(1)

where T is the total emergy, $E_j$ is the available energy or exergy, and $T_{r_j}$ is the unit emergy value.

2.2.2. LCA-Emergy Element Calculation Model

To determine the emergy of the whole life cycle of the building system, seven types of inputs are considered and analyzed: solar, mass, electricity, water, diesel fuel, and gasoline.

(1)

Solar irradiation model

The solar emergy can be used in Equation (2):

$$E_{Solar}=A\times E_J\times (1-\omega )\times t_c\times T_{UEV\mathrm{s}}$$

(2)

where $E_{Solar}$ represents the solar emergy in the construction process, A is the site surface, $E_J$ is the solar radiation amount (3.5 × 10⁹ J/m²), ω is the surface albedo (0.7), $t_c$ is the construction time, and $T_{UEV\mathrm{s}}$ is the unit emergy values.

(2)

Mass Calculation Model

The mass of the emergy calculation model equation is calculated as follows (3).

$$E_{\mathrm{M}}={\displaystyle \sum _{i=1}^n{Q}_i}\times T_{U\text{EVs-mass}}$$

(3)

where $E_{\mathrm{M}}$ is the mass emergy; $Q_i$ is mass quantity, and $T_{U\text{EVs-mass}}$ is the unit emergy value.

(3)

Electricity calculation model

The electricity calculation equation can be obtained by using Equation (4):

$$E_{\mathrm{ele}}=M_{\xi }\times T_{U\text{EVs-ele}}$$

(4)

where $E_{\mathrm{ele}}$ is the emergy of electricity in the building system; $M_{\xi }$ is the electricity quantity; and $T_{U\text{EVs-ele}}$ is the unit emergy value.

(4)

Water emegy model

The water emergy has two parts. Firstly, it should be calculated in the building demolition and construction stage. The specific Equation (5) can be used as follows:

$$E_{water}=V\times \rho \times G\times UE{V}_w$$

(5)

where $E_{water}$ is the water emergy, $V$ is the water volume, $\rho$ is the water density, G is the Gibbs energy of water (4.92 J/g), and $UE{V}_w$ is the water transformity.

Secondly, the water emergy should also be considered in the operation phase, and the equation can be utilized as (6).

$$F_{water}=V_o\times N_o\times T_o\times \rho \times G\times UE{V}_w$$

(6)

where $F_{water}$ is the water emergy in the building operation stage, $V_o$ is the water volume per day for one person (25 L/d/p), $N_o$ is the employee number (the number is 200), and $T_o$ is the working time (280 days in this study).

(5)

Diesel fuel emergy model

Because of the machinery used, diesel fuel is necessary for the building system. The equation can be calculated in Equation (7):

$$E_{\mathrm{diesel}}=\varsigma \times \zeta \times UE{V}_d$$

(7)

where $E_{\mathrm{diesel}}$ represents the emergy of the diesel fuel, $\varsigma$ is the amount of diesel oil used in the buildings system, $\zeta$ shows the calorific value of diesel fuel, and $UE{V}_d$ is the unit emergy value.

(6)

Gasoline emergy calculation model

It can be computed based on Equation (8):

$$E_{gasoline}=\lambda \times \upsilon \times UE{V}_g$$

(8)

where $E_{gasoline}$ is the gasoline emergy, $\lambda$ is the gasoline quantity, $\upsilon$ is the calorific value, and $UE{V}_g$ is the unit emergy value.

(7)

Human labor emergy calculation model

The emergy of human labor can be counted from Equation (9):

$$E_{Human}=T_{work}\times N_P\times W_d\times UE{V}_H$$

(9)

where $E_{Human}$ is the emergy of human labor, $T_{work}$ is the working time (8 h), $N_P$ is the number of employed workers, $W_d$ is the working day, and $UE{V}_H$ is the unit emergy value.

2.2.3. Emergy Diagram

In Figure 2, the building system's whole life cycle energy diagram is designed and defined in detail.

There are four primary parts, including renewable resource input (on the left), non-renewable inputs (on the top), main system boundary (in the middle), and the external system (on the right). The whole life cycle of a building system consists of five stages, namely the building material stage, material transformation stage, construction stage, operational stage, and demolition stage, respectively.

2.2.4. Sustainable Indicators

For sustainability assessment in the building system, a series of indexes have been adopted to demonstrate sustainable hierarchy, as follows:

(1)

Emerge intensity (E_p) represents the emergy per person.

(2)

Emergy per RMB (E_e) demonstrates the amount of emergy per unit of money.

(3)

Emergy density (E_d) illustrates the emergy per unit area.

(4)

Renewability rate (R_e) explains the proportion of renewable inputs.

(5)

Non-renewability rate (N_r) displays the proportion of non-renewable inputs.

(6)

Non-renewability rate of purchased resource (N_p) shows the ratio of non-renewable resources purchased.

(7)

Purchased emergy dependence level (P_e) describes the proportion of resources purchased by external inputs.

(8)

Emergy investment ratio (EIR) instructs emergy ratio of external investment.

(9)

Environmental loading ratio (ELR) represents the system’s load pressure on the environment.

(10)

Emergy yield ratio (EYR) expresses the emergy independent level in the system.

(11)

Emergy sustainability index (ESI) states the sustainability degree of the target system. According to the standard [41], the basic criteria is ESI = 1(if ESI < 1, unqualified).

According to the study of architectural cases, the selection criteria of sustainability goals include three types: basic criteria (average emergy index group), procedural criteria (basic emergy index group), and final qualitative criteria (comprehensive key emergy index).

Basic criteria can express the basic structural sustainability of the building, including Emerge intensity (E_p), Emergy per RMB (E_e), Emergy density (E_d), etc.

Procedural criteria demonstrate the sustainable process of building cases involving Renewability rate (R_e), Non-renewability rate (N_r), Purchased emergy dependence level (P_e), and Emergy investment ratio (EIR).

The final qualitative criteria provide the comprehensive sustainability assessment results based on the Environmental loading ratio (ELR), Emergy yield ratio (EYR), and Emergy sustainability index (ESI).

Among them, the parameters positively correlated with the sustainability effect are R_e, P_e, EIR, and ESI, respectively. The rest are negative feedback indicators, and the larger the value, the lower the sustainability of the building system (Figure 3). According to the logical, in-depth analysis of indicators, these indicator groups can be divided into two categories; one is the basic indicators (E_p, E_e, E_d, R_e, N_r, N_p, P_e), the other is the comprehensive key indicators (EIR, ELR, EYR, ESI).

EIR shows the economic competitiveness of building systems. ELR, as the definition, represents the environmental load concentration of a building system, which is a high standard, when it is greater than 10. EYR explains the building system’s emergy generation level and the relationship's closeness with the external environment. ESI can be counted based on EYR and ELR, which elucidates the final sustainability degree of the building system. In general, it is unsustainable when the value is less than 1 [41].

2.2.5. Sensitivity Analysis

In order to obtain an accuracy assessment result, the sensitivity analysis needs to be realized. In this paper, the sensitivity analysis has been executed for the main impact element in the building system. In general [23], the calculated equation can be used as follows:

$$E_p(i)=[(E+\alpha \times i)]\times [(UEV+\beta \times i)]$$

(10)

Herein, $E_p$ represents the emergy, E is the energy (J), mess (kg), and money ($), respectively, UEV is the unit emergy value, $\alpha$ shows the emergy error, and $\beta$ is the UEV error.

2.3. Artificial Neural Network (ANN) Method

Artificial neural network (ANN) is widely adopted as a study method in system prediction evaluation, which has rapidly developed in the last twenty years [42,43]. The ANN structure generally consists of three layers of neurons: the input layer, the hidden layer, and the output layer [44,45]. In Figure 4, an integrated prediction model has been designed to be coupled based on the LCA-Emergy view.

At the bottom level of the design drawing is the input of a series of indexes, including basic and comprehensive indicators. The intermediate structure is the hidden layer and the various feedback inputs, involving an unsupervised training model and material flow, energy flow, and information flow feedback. In the top-level output, the sustainability state of the entire building system is generated in the long run.

The basic mathematical formula for neural networks can be found in Equations (11) and (12).

$${\varphi }_{\partial }={\displaystyle \sum _i^m{A}_i\times Y_i}$$

(11)

$$T_{\mu }=\mu \times (C_{\mu }+D_{\mu })$$

(12)

where ${\varphi }_{\partial }$ represents the sample, $Y_i$ shows the input vectors, and $T_{\mu }$ is the output vectors. $\mu$ and $C_{\mu }$ demonstrate the bias vectors. $D_{\mu }$ denotes the threshold value.

2.4. Case Study

For the sake of regional economic development and personnel exchange, this new airport was built in Xuancheng city, which is located between 29°57′ and 31°19′ north latitude and 117°58′–119°40′ east longitude and belongs to the Anhui Province. The airport adopts an A2 class general airport level to design and build [46], mainly used for aviation forest protection, disaster protection, medical rescue, transportation, flight training, tourism, and other flight activities.

The airport building complex consists of three categories, namely the airport terminal, dormitory, and hangar, which are all reinforced concrete structures (Figure 5), with a total investment estimated at 228.83771 million yuan. Therein, the airport complex building has 6810 square meters of construction area (four floors) and covers an area of 1966 square meters. Correspondingly, the data for the hangar (two floors) and dormitory complex building (four floors) are 5600 m² and 4880 m², 6207 m² and 2126 m², respectively. The study case is from the machinery Industry Sixth Design and Research Institute Co., Ltd. and was authorized.

3. Results and Discussion

As the main outcome and discussion session, five subsections are displayed: LCA-Emergy analysis, sustainable indicators analysis, sensitivity analysis, unit emergy value, and ANN-Emergy analysis.

3.1. LCA-Emergy Analysis

In Section 3.1, five stages of the building system have been selected and analyzed (Figure 6), which are the building materials stage, materials transportation stage, construction stage, operation stage, and demolition stage, respectively. Based on the LCA-Emergy view, the emergy of building material production and operational stage plays a key role, accounting for roughly 92.4% of the total emergy in the long run, which is the primary contributor according to 30 years of service life of an airport building [47].

By evaluating each phase separately, the proportion of the emergy at the operation stage is the largest (61.64%), followed by the building material stage (30.49%), the demolition stage (3.7%), the construction stage (2.31%), and the material transportation stage (1.52%). The development trend can be seen in Figure 6.

From the building material point of view to analyze singly, the largest amount of emergy is concrete, followed by cement and steel, accounting for 60.7%, 20.3%, and 8.65%, respectively. The detailed trend has been shown in Figure 7A,B. From the perspective of a device (in Figure 7C,D), the first three inputs are sorted as pressure welding machines (33.8%), followed by butt welding machines (25%), and tower cranes (11.3%).

3.2. Sustainable Indicators Analysis

There are three categories of indexes in Section 3.2, which are the average emergy index group, basic emergy index group, and comprehensive key emergy index group.

Firstly, three kinds of emergy average indicators are presented: Emergy intensity (E_p) is 1.51 × 10¹⁷ sej/unit, which is a high degree. Emergy per RMB (E_e) is 3.19 × 10¹² sej/unit, showing better sustainability. Meanwhile, Emergy density (E_d) has a sustainable state, and the value is 8.95 × 10¹⁵ sej/unit.

Secondly, basic emergy indicators consist of four inputs, which are Renewability rate (R_e), Non-renewability rate (N_r), Non-renewability rate of purchased resource (N_p), and Purchased emergy dependence level (P_e), respectively. R_e is 0.23, which illustrates a poor ecological level and needs to supplement to enhance the sustainability level. N_r is 0.77, demonstrating a worse ecological extent. N_p is 0.32, which should minimize the purchased resource emergy input. P_e is 3.13, and it is not enough to improve the competitiveness of the building system.

In the last, the Emergy investment ratio (EIR), Environmental loading ratio (ELR), Emergy yield ratio (EYR), and Emergy sustainability index (ESI) were counted and analyzed. EIR is 0.23, instructing the economic input degree of the building system is weak. ELR is 3.423, interpreting the ecological load degree (medium intensity). EYR is 2.291, demonstrating the building system’s poor ability to generate emergy. As the vital indicator, ESI has been calculated based on EYR and ELR; it is 0.669, which is unsustainable in the long run according to the standard.

3.3. Sensitivity Analysis

Sensitivity analysis is closely related to the main influencing factors, which will greatly affect the accuracy of the final result. This study has six primary contributors: concrete, cement, steel, pressure welding machines, followed by butt welding machines and tower cranes, etc. To achieve the sensitivity calculation quantitatively, a hypothesis has been executed: each of the main emergy inputs changes by 10%, and other parts remain constant to verify the indicator’s sensitivity variation. Figure 8 shows the final sensitivity results of six main contributors, which illustrates the indicators change trend after conducting the hypothesis (Figure 8A–E).

From Figure 8A,B, the pressure welding machine has the most significant fluctuation (6.019%), followed by the tower crane (5.427%) and concrete (5.113%). Simultaneously, Figure 8C–E presents the indicators variations between the former and the latter. The order changes from largest to smallest is ESI (7.72%), E_e (5.76%), E_p (5.3%), EYR (3.5%), N_p (3.03%), ELR (2.81%), EIR (2.13%), N_r (1.28%), E_d (0.72%), and P_e (0.48%). Therein, Re (−5%) is distinctly different from others, which tends to decrease.

3.4. Unit Emergy Value (UEV)

As the vital analysis part, unit emergy value (UEV) is the core content from the view of the emergy method. It is one of the key indicators to evaluate the sustainability of the building system, which can judge the health state of the entire building system. To date, in the field of building systems, not all building systems have UEV analysis when the emergy approach is used.

To find out the current research status of UEV clearly, relevant kinds of literature in the last five years have been combed and sorted in

Table 1. UEVs literature review list.

Year	Author	Baseline	UEV	LCA View	Emergy View	ANN Perspective	Country	Ref.
2017	Hwang et al.	New	×	√	×	×	USA	[30]
2017	Jae and William	New	×	×	√	×	USA	[31]
2020	Thomas and Praveen	New	×	×	√	×	India	[26]
2021	Wenjing et al.	None	×	√	√	×	China	[35]
2021	Suman et al.	None	×	×	√	×	USA	[24]
2022	Xinnan et al.	New	√	√	√	×	China	[36]
2023	Junxue et al.	New	√	√	√	√	China	This study

.

Through the whole building system based on emergy data, the UEV of airport buildings has been estimated (6.53 × 10¹⁵ sej/m²). By comparing with the UEV data in Table 1, the value is relatively sustainable, which is consistent with the sustainability parameter of ESI. For instance, take the latest emergy study as the case to analyze; the UEV of the case [36] is 2.14 × 10¹⁸ sej/m², higher than that studied in this paper, elucidating that the design of building inputs in this paper is feasible.

3.5. ANN-Emergy Analysis

In addition to calculating and evaluating the existing sustainability effects in the building system, the future ecological sustainability of the building system also needs to be investigated and forecasted. In this study, taking Emergy Sustainability Indicator (ESI) as the pivotal indicator, the artificial neural network (ANN) method has been used to predict the life cycle of the building. In Figure 9A, the best validation performance is 4.74 × 10⁻⁵ at epoch 30. Figure 9B demonstrates the error distribution; the zero error is at the −0.3044 station. The regression analysis graphs have been computed in Figure 9C and show the regression accuracy of training, validation, testing, and total, which are 0.99989, 0.99998, 0.99991, and 0.99989, respectively. Through the regression accuracy, the robustness of ANN has a qualified form for the building system. In Figure 9D, the uncertainty percentage of all factors has been illustrated, which demonstrates the fluctuation range.

4. Improved Sustainable Design

Two measures have been provided to improve the sustainable effect in the building system, including clean energy embedded design and a coupling design of landscape subsystem.

4.1. Clean Energy Embedded Design

The use of renewable energy affects the sustainability of building systems, according to the study in this paper (Figure 10). Up to now, solar and wind power are the types of clean energy the Chinese government has heavily promoted, which have been investigated by many scholars [48,49].

In Figure 10, a clean, comprehensive model of integrating solar energy and wind power has been designed and explored to confirm the sustainability change. Figure 10 consists of three sections: the composite clean energy system, the feedback system, and the building system, respectively. The composite clean energy system relies on a range of devices to supply electricity. Then, through the feedback system, the stability of the power system is guaranteed and translated into the electricity consumption mode of the building system.

In this study, three scenarios are presented and analyzed, including 5%, 10%, and 20% of renewable energy replacement, respectively. Through the calculation of three key indicators (ELR, EYR, and ESI), changes in the three indicators are calculated and displayed in Figure 11. Figure 11A–D depicts the sustainable index changes. In Figure 11A–C, the indicators in the three scenarios were compared with the original indicators, showing that the environmental load rates of the entire building system are reduced (from 3.423 to 3.028/2.561/2.315), which are positive roles for the entire building system. Similarly, at the same time, the emergy production rate also has a similar trend (from 2.291 to 2.182/1.972/1.882). However, the reduced energy production rate has a negative effect on the sustainability of the building system. Based on ELR and EYR, ESI has been evaluated quantitatively, which are 0.669 of original data, 0.721 for the 5% scenario, 0.77 for the 10% scenario, and 0.813 for the 20% scenario, respectively (Figure 11D). Analyzing the trend as a whole, the sustainability parameter (ESI) is a positive trend, demonstrating that the sustainability of the entire building system is raised with the continuous input of clean energy. The fact is verified that using clean energy can optimize the sustainability level of building systems based on the LCA-Emergy method.

Nevertheless, in practice, several disadvantages hinder their promotion and application [50], such as enormous investment, professional and technical barriers, and geographical conditions. In this context, financial subsidies and favorable tax policies are effective measures to increase the use of clean energy and improve the building system’s sustainability.

4.2. A Coupling Design of Landscape Subsystem

The coupling of landscape ecosystems contributes to the mitigation of environmental climate change. In this study, three types of elements are utilized to design landscape systems, including trees, water, and flowers. Four landscape design schemes were designed based on these basic elements, which will be coupled to the building system to enhance sustainability. In Figure 12, the basic implementation framework is shown.

In Figure 13, four design schemes of landscape subsystems have been displayed about the sustainability effect. Based on the critical indicator group (ELR, EYR, and ESI) displays, a clear pattern can be obtained from Figure 13a–d. When the architectural system is coupled with the landscape design subsystem, sustainability is significantly improved, no matter what kind of landscape subsystem input.

However, disparate types of landscape design reveal distinct differences in the sustainable improvement of the building system. Figure 13b has the best sustainability improvement, followed by Figure 13a,c,d. In Figure 13b, the primary feature is the water landscape design, which demonstrates that the water landscape has a relatively ameliorative effect on sustainability. In Figure 13c, green space design is the major characteristic (large area than others’ design schemes), showing better sustainability (like the effect of water landscape design). In Figure 13d, a landscape design scheme that integrates green space and trees is more sustainable than a single tree planting (Figure 13a). To sum up, the sustainability effects of landscape design elements are water landscape design, green space design, green tree design, and flower design in order (from good to bad). The above facts also verify the positive sustainability impact of landscape subsystems on building systems, which gives architects new design ideas and inspiration to enhance the sustainability of building systems.

5. Conclusions

To investigate the environmental sustainability of the building system, LCA-Embodied energy (Emergy)-ANN methodology was implemented to analyze the airport buildings in this study.

Through LCA-emergy calculations, five stages of the building system have been compared. The building operation stage has the highest emergy mass, approximately 62% of the total emergy, followed by the building material stage (30.49%), the demolition stage (3.7%), the construction stage (2.31%), and the material transportation stage (1.52%), which identifies priorities if it is necessary to reduce the emergy level of the entire building system.

Simultaneously, a series of indicators have been selected and conducted to reveal sustainability grade. In particular, the Emergy sustainability index (ESI = 0.67) provides the final unsustainable state, which should consider improvement measures. In the long run, an artificial neural network (ANN) was utilized to predict the ESI floating range of five stages, which validates again the dominance of the building operation phase and the building material production phase.

To improve the environmental sustainability of the entire airport building, clean energy was embedded in the building system by replacing non-renewable energy sources. At the same time, the landscape subsystems were also coupled to the building system to promote a sustainable hierarchy. The above two strategies positively affect the building system's sustainability, providing references for architects and civil engineers.

In this study, several drawbacks also limit further development. For example, as a complex project, data acquisition of building systems is limited to some extent. It is not possible to collect 100% of the overall construction phase data, which has a negative impact on the study results. At the same time, for the building system, since the building is in the normal use stage, the investigation of the building operation stage and the building demolition stage needs to be predicted to a certain extent according to the daily operation data, which also influenced the results of the study. The future research focus of this study requires data tracking of building systems to verify the accuracy of the results of this study.