Selection of digital fabrication technique in the construction industry—A multi-criteria decision-making approach

Abstract

Digital fabrication techniques, in recent decades, have provided the basis of a sustainable revolution in the construction industry. However, selecting the digital fabrication method in terms of manufacturability and functionality requirements is a complex problem. This paper presents alternatives and criteria for selection of digital fabrication techniques by adopting the multi-criteria decision-making technique. The alternatives considered in the study are concrete three-dimensional (3D) printing, shotcrete, smart dynamic casting, material intrusion, mesh molding, injection concrete 3D printing, and thin forming techniques. The criteria include formwork utilization, reinforcement incorporation, geometrical complexity, material enhancement, assembly complexity, surface finish, and build area. It demonstrates different multi-criteria decision-making techniques, with both subjective and objective weighting methods. The given ranking is based on the current condition of digital fabrication in the construction industry. The study reveals that in the selection of digital fabrication techniques, the criteria including reinforcement incorporation, build area, and geometrical complexity play a pivotal role, collectively accounting for nearly 70% of the overall weighting. Among the evaluated techniques, concrete 3D printing emerged as the best performer, however the shotcrete and mesh molding techniques in the second and third positions.

Graphical abstract

Keywords

digital fabrication / multicriteria decision-making / concrete 3D printing

1 Introduction

Digital fabrication (DF) has emerged as a sustainable and effective technique in the construction, automobile, medical, and other industries. Digital manufacturing in construction is a layer-by-layer deposition process, which allows the fabrication of complex and structurally efficient structures with minimal negative environmental impacts. Some examples of DF in construction projects involve concrete three-dimensional (3D) printed bridges [1–7], 3D printed houses [8,9], pavilion [10,11], mesh molding double curve walls [12–14], smart dynamic casting mullions [15], eggshell columns and piers [16,17], and other structural components. The construction industry employs different types of DF techniques based on the type of structure and its requirements. The effectiveness and sustainability of the structure primarily depends on the selection of appropriate DF technique. The advantages and limitations of every DF technique probably vary based on the parameters, including reinforcement incorporation, geometrical complexity, surface quality, and assembly complexity. Prior to the selection of the appropriate DF technique, the complexity of the element and the necessary parameters need to be considered to reduce negative environmental impact without affecting efficiency. The selection of the DF technique then results in projects becoming more economical and environment-friendly.

DF techniques have some limitations such as anisotropic behavior, a complication in reinforcement incorporation, as well as deficiency in structural integrity, surface finish, and build area limitations [18–22]. Researchers [23] have critically reviewed the selection of reinforcement, assembly, and formwork systems among different DF techniques in the construction industry.

Concrete 3D printing is a rapidly growing DF technology in the construction industry [22,24,25]. The layer-by-layer process increases the anisotropic behavior because of the interface bond between the two layers [26–29]. Anisotropic behavior can be observed in all DF techniques except smart dynamic casting [30]. The strength of concrete 3D printing structures is influenced by various parameters including nozzle geometry, printing speed, and open time [31–33]. Tay et al. [34], have designed functionally graded material for topology-optimized structures by altering the printing speed. Results have shown that the optimized structure has strength-to-weight ratio that is 50% higher than the traditional construction. Microfibers have been used in the material intrusion method for the fabrication of a 12 m-long foot-over bridge [35]. Burger et al. [16] have used conventional reinforcement in thin formwork for the Future Tree project. Similarly, Hansemann et al. [36] have also used conventional reinforcement, in shell and fill, for a smarter ceiling system. Salet et al. [7] have utilized cable reinforcement for a bicycle bridge by developing a custom nozzle after exploring the behavior of backflow and downflow nozzles. Moreover, Kazadi et al. [37] have reviewed the current potential and challenges of the reinforcement strategy. They recommended the combination of micro cable reinforcement and U-nail reinforcement to enhance tensile performance without affecting interlayer bonding. Tuvayanond and Prasittisopin [38] have conducted a comprehensive review of Design for Manufacturing and Assembly in DF, exploring its advantages. Vantyghem et al. [39] have made a concrete 3D printed girder with 18 segments by using post-tensioning reinforcement along with conventional reinforcement combined with shell and fill DF technique to ensure structural integrity. Similarly, the post-tensioning reinforcement used in bicycle bridge, branching column construction [17], 3D printed stairs [40], and Nijmegen Bicycle Bridge [41] to ensure structural integrity while adopting a segmental printing process. On the other hand, Hadid et al. [2], Ma et al. [5], and Burry et al. [42] have designed and printed bridges using arch design without reinforcement, which requires formwork utilization during the assembling process. The seven criteria are interlinked, for instance, inadequate build area or geometric complexity leads to assembly. Therefore, during the selection process of DF techniques, the criteria need to be considered and the weightage of each criterion differs as per the alternatives and the nature of requirements. The interlinking multicriteria increase the complexity in the selection of DF technique. To resolve this problem, multi-criteria decision-making (MCDM) techniques are used to decide the selection of DF technique.

Several researchers have investigated the use of the MCDM technique in the field of additive manufacturing. For instance, Raja and Rajan [43] have used the technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for selecting suitable Fused Deposition Modeling (FDM) machines for Indian non-government organizations. In this case, nine criteria—price, build volume, temperature, extruder type, printing speed, material, safety, geometric accuracy, and environmental factor—are used to determine the FDM printer. On the other hand, Chodha et al. [44] have adopted TOPSIS with an objective weighting method (ENTROPY) to select an industrial arc welding robot. Palanisamy et al. [45] have used MCDM for two stages of the selection process. In stage one, the additive manufacturing machine was selected. Subsequently, the best-worst method could determine the suitable material based on the respondent’s requirements. Agrawal [46] adopted TOPSIS, Simple Additive Weighting (SAW), Multi-Objective Optimization based on Ratio Analysis (MOORA), and Vlsekriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methods for the selection of sustainable materials in three different additive manufacturing techniques. Fifteen, ten, and seven criteria are considered for the material selection of FDM, SLS, and SLA-based printers, respectively. MCDM techniques are widely used in the fields of additive manufacturing for a selection process of methods, materials [45–50], etc. However, in the field of construction, the adoption of MCDM is relatively less as compared to DF in other disciplines. Iranfar et al. [51] have studied the selection of binders for construction of a human habitat on Mars. They used three different MCDM techniques, namely, TOPSIS, VIKOR, and Weighted Aggregated Sum Product Assessment (WASPAS) with the hybridization of fuzzy Analytic Hierarchy Process (AHP). To determine the most suitable binder from eight alternatives (i.e., geopolymer, ordinary Portland cement, sulfur cement, sintered material, calcite precipitation, polymer-bound regolith, geo-thermite reactions, and regolith-based magnesium oxychloride) they considered 14 criteria: shipping, strength, durability, recyclability, sustainability, working condition, ambient temperature, curing time, additives, energy requirement, cosmic shield, safety, water requirement, and availability. The result showed that shipping was the highest weighted influencing criterion and geopolymer cement was the best material among the other alternatives for construction on Mars. Yoris-Nobile et al. [49] have determined the performance of 3D printed low-clinker cement mortar and geopolymers using MCDM (i.e., TOPSIS and WASPAS methods) and Life cycle assessment (LCA). The dosage for 3D printing mortar was selected using MCDM, accounting for material cost, printability, strength, and LCA. As a result, due to low negative environmental impact and material cost, the low clinker cement mortar secured the top ranking among other alternatives. In addition to that, Miç and Antmen [52] have used TOPSIS, WASPAS, and MOORA methods for the university location selection process.

Research Gap: A potential research gap is the lack of effective utilization of the MCDM approach for the selection of the DF method in the construction industry, considering various criteria and alternatives.

Research Questions:

1) How can the MCDM approach be effectively applied to select a DF method in the construction industry, while prioritizing high reinforcement incorporation, the ability to produce complex geometrical elements in a large build volume, and minimal assembly complexity without a conventional formwork?

2) How can the selection of technique reduce negative environmental impact while achieving low cost with minimal construction timing? This research question involves exploring different MCDM techniques and their application to evaluate and rank DF methods based on various complex criteria such as formwork utilization, reinforcement incorporation, geometrical complexity, material enhancement, assembly complexity, surface finish, and build area.

This paper presents the selection of DF techniques using different alternatives and criteria, by adopting the MCDM technique. Three different MCDM techniques such as TOPSIS, WASPAS, and MOORA are demonstrated using both subjective and objective weighting methods. Moreover, a pairwise comparison was made among the DF techniques.

The research methodology is outlined in Section 2, with two pivotal subsections. Subsection 2.1 explores the weighting of criteria, a crucial aspect of the MCDM process, elucidating how the relative importance of various factors is determined. Subsection 2.2 describes the MCDM technique employed in this study, by providing a comprehensive understanding of the analytical framework utilized for the evaluation of alternatives. Section 3 focuses on the study parameters, in two key sub-sections. Subsection 3.1 addresses the alternatives considered for assessment, while Subsection 3.2 outlines the process employed in the selection of evaluation criteria. Section 4 presents the results and initiates a detailed discussion. Subsection 4.1 sheds light on the sensitivity analysis and offers insight into how variations in criteria weights influence the overall rankings. Section 5 encapsulates the key findings and implications drawn from the study.

2 Methodology

The MCDM technique involves several steps [53]: 1) problem identification; 2) defining the requirements; 3) establishing the goals; 4) alternative identification; 5) criteria development; and 6) identifying and applying MCDM techniques, as illustrated in Fig.1.

Problem Identification. Selecting a DF method based on manufacturability and functionality requirements.

Defining the requirement. Selection of DF method with a high choice of reinforcement incorporation, high capability to produce complex geometrical elements in large build volume, and minimal assembly complexity without conventional formwork.

Establishing the goals. 1) Selection of DF method satisfying the requirements. 2) Establishing the pairwise comparison among the DF methods to determine the place to leverage each DF method to meet the level of requirements.

Identifying and applying MCDM techniques. This paper adopts objective and subjective weighting approaches for decision-making. The advantages of the AHP weighting method are the ability to incorporate qualitative and quantitative factors, its hierarchical formulation, cost-effectiveness, time efficiency, precise results, expert-based ranking, flexibility for re-arrangement, and utilization of pairwise comparison matrices [54–56]. Therefore, in this study, the AHP method was used for subjective weighting. The Entropy method was used as an objective weighting method since it can measure the consistency and dispersion of data. Moreover, it provides reliable and stable weightings for decision criteria, enhancing not only the objectivity but also the robustness of the decision-making process [57–59].

The MCDM analysis conducted in this study employed three distinct techniques, namely TOPSIS, WASPAS, and MOORA. These three MCDM techniques were chosen for their demonstrated effectiveness in evaluating complex decision-making scenarios, particularly in the context of selecting the most suitable DF technique for construction (Fig.2). TOPSIS [60] excels in identifying alternatives that strike a balance between proximity to the positive ideal solution and remoteness from the negative ideal solution, providing a comprehensive evaluation framework. WASPAS [61] offers a flexible approach by allowing the integration of subjective and objective criteria, facilitating a nuanced assessment. MOORA [62], known for its versatility in handling multiple objectives, was selected to ensure a thorough evaluation considering various dimensions of performance.

These chosen techniques collectively form a robust analytical framework enabling a thorough and systematic evaluation of alternatives. Its combined application ensures that both subjective insights and objective assessments are considered, resulting in a well-informed selection of the most suitable DF technique in the construction industry.

Fig.3 shows the subjective and objective weighting approaches for MCDM. Both the subjective and objective weighting approaches are utilized with the three most used MCDM techniques in the field of additive manufacturing [45–50]: TOPSIS [60], WASPAS [61], and MOORA [62]. By applying these MCDM techniques, the criteria are sorted based on the requirement and ranking of the DF technique using the normalized score.

2.1 Weighting of criteria

2.1.1 Analytic hierarchy process method

The AHP weighting method [63] is a subjective method, driven by a pairwise comparison process and purely influenced by the judgments of decision-makers. The Preference evaluation matrix [P] is made by N × N elements; seven criteria are applied, hence n = 7. For instance, while considering the criteria C₁ and C_2, the comparative evaluation is expressed in the importance scaling as shown in Tab.1. The relative importance is converted into numerical scales C₁₂ and C₂₁, where, the comparative evaluation of C₂ and C₁ (C₂₁) is the reciprocal of the comparative evaluation of C₁ and C₂ (C₁₂), i.e., C₂₁ = 1/C₁₂. The preference evaluation matrix is shown in Tab.2.

2.1.2 Entropy method

The entropy weighting method [64] works as an objective method. It is an unbiased data method requiring the following steps.

Step 1: Decision Matrix

The decision matrix [X] is generated using Eq. (1). Column (i) denotes alternatives, and row (j) represents criteria.

(1)\begin{array}{c}{X}_{ij}=\left[\begin{array}{ccc}{x}_{11}& \cdots & x_{1n}\\ ⋮& & ⋮\\ x_{m1}& \cdots & x_{mn}\end{array}\right]\text{.}\end{array}

Step 2: Normalized Matrix

The decision matrix is normalized using Eq. (2).

(2)f_{ij}=\frac{X_{ij}}{{\sum }_{j=1}^n{X}_{ij}}\text{,}

where f_{ij} is the normalized value of the decision matrix.

Step 3: Normalized Entropy Value

The normalized entropy value is calculated using Eq. (3).

(3)E_j=-K\sum _{i=1}^m{f}_{ij}\mathrm{l}\mathrm{n}{f}_{ij}\text{,}

where K = 1/lnm and m is equal to a number of alternatives.

Step 4: Entropy objective weights

The entropy objective weightings (W_j) of the alternatives are determined using Eq. (4).

(4)W_j=\frac{1-E_j}{\sum _{j=1}^n\left(1-E_j\right)}

where the sum of the objective weights (W_j) for entropy is equal to one.

2.2 Multi-criteria decision-making techniques

2.2.1 Technique for order of preference by similarity to ideal solution

Step 1: Decision Matrix

The decision matrix [X] is generated using Eq. (1).

Step 2: Normalized decision matrix

The normalized standard decision matrix (r_{ij}) is calculated using Eq. (5). Ranking

(5)r_{ij}=\frac{X_{ij}}{\sqrt{\sum _{j=1}^m{X}_{ij}^2}}\text{.}

Step 3: Weighted normalized decision matrix

The weighted normalized matrix is determined by multiplying weight (W_j) with normalized standard decision matrix (r_{ij}) as shown in Eq. (6).

(6)v_{ij}=W_j{r}_{ij}.

Step 4: Best and worst ideal solution

The best and worst ideal solutions can be determined using Eq. (7) and (8), respectively.

(7)v^{+}=\left\{\left(\sum _i^{max}{v}_{ij}/j\in J\right),\left(\sum _i^{min}{v}_{ij}/j\in J^{\prime }\right)\right\}\text{,}

(8)v^{-}=\left\{\left(\sum _i^{min}{v}_{ij}/j\in J\right),\left(\sum _i^{max}{v}_{ij}/j\in J^{\prime }\right)\right\},

where J represents beneficial attributes and J^{\prime } represents non-beneficial attributes.

Step 5: Separation distance between the best and worst ideal solution

The separation distance between the best ideal solution is determined using Eq. (9) and the worst ideal solution is calculated by Eq. (10).

(9)S_i^{+}=\sqrt{\sum _{j=1}^n{\left(v_{ij}-v_j^{+}\right)}^2},

(10)S_i^{-}=\sqrt{\sum _{j=1}^n{\left(v_{ij}-v_j^{+}\right)}^2}.

Step 6: Relative closeness

The relative closeness (C_i) of the ideal solution is calculated using Eq. (11).

(11)C_i=\frac{S_i^{-}}{S_i^{+}+S_i^{-}}.

Step 7: Ranking

The alternatives are ranked according to the relative closeness value (C_i).

2.2.2 Weighted aggregated sum product assessment

Step 1: Decision Matrix

The decision matrix [X] is generated in the TOPSIS method, as shown in Eq. (1).

Step 2: Normalized decision matrix

The normalized standard decision matrix is calculated as per the nature of the criteria. If beneficial criteria are calculated using Eq. (12), the non-beneficial criteria are determined by Eq. (13).

(12){\overline{x}}_{ij}=\frac{x_{ij}}{\mathrm{m}\mathrm{a}\mathrm{x}\left(x_{ij}\right)},

(13){\overline{x}}_{ij}=\frac{\mathrm{m}\mathrm{i}\mathrm{n}\left(x_{ij}\right)}{x_{ij}}.

Step 3: Weighted normalized decision matrix

The Weighted normalized decision matrix is determined by the weighted sum method (WSM) as shown in Eq. (14).

(14)Q_i^{\left(1\right)}=\sum _{j=1}^n{\overline{x}}_{ij}{W}_j.

The Weighted normalized decision matrix is determined by the weighted product method (WPM) as shown in Eq. (15).

(15)Q_i^{\left(2\right)}=\prod _{j=1}^n{\left({\overline{x}}_{ij}\right)}^{W_j}.

Step 4: Improvement of ranking accuracy

The WSM and WPM values are combined to improve the ranking accuracy with help of generalized criteria as shown in Eq. (16).

(16)Q_i=\lambda \cdot Q_i^{\left(1\right)}+\left(1-\lambda \right)\cdot Q_i^{\left(2\right)},

where \lambda varies from 0 to 1. To give equal preference to WSM and WPM the \lambda is considered as 0.5.

2.2.3 Multi-objective optimization based on ratio analysis

Step 1: Decision Matrix

The decision matrix [X] is generated as in the TOPSIS method, as shown in Eq. (1).

Step 2: Normalized decision matrix

(17)x_{ij}^{\ast }=\frac{x_{ij}}{{\left(\sum _{i=1}^m{x}_{ij}^2\right)}^{1/2}}.

Step 3: Assessment value

(18)y_i=\sum _{j=1}^g{W}_j{x}_{ij}^{\ast }-\sum _{j=g+1}^n{W}_j{x}_{ij}^{\ast },

where g represents number of beneficial criteria, y_i represents the performance score.

3 Study parameters

3.1 Alternatives

Different types of DF techniques are used in the field of construction and other industries, and can be classified as either material extrusion or material intrusion methods. Concrete 3D printing, shotcrete, mesh molding, thin forming, smart dynamic casting, and injection concrete 3D printing are based on material extrusion. In general, two types of material intrusion techniques are adopted, namely selective paste intrusion and selective binder activation. DF techniques are capable of manufacturing complex optimized structural components with minimal material usage and without conventional formwork. Development of DF techniques has been driven by features such as dealing with topology-optimized structures and lightweight structures, and the techniques also help to overcome the problem faced in the conventional casting methods, such as the need for complex and costly formwork [65,66].

3.1.1 Concrete three-dimensional printing

Concrete 3D printing is a layer-by-layer printing process, which works based on the material extrusion method as shown in Fig.4. The concrete for 3D printing needs to have specific properties including extrudability, printability, and buildability properties to attain high-quality printing with minimal geometric deviation [67–70]. Khoshnevis [71] used the trowel system (i.e., contour crafting) to achieve high surface quality. The type of printer employed for concrete 3D printing is based on the build area requirement and complexity of the structure. In general, robotic arm, polar, framed, and delta-type printers are used for small-scale printing [36,39,72,73], and for large-scale printing crane and cable printer are used [74–76]. Gantry printers are used for both small and large-scale printing [77].

3.1.2 Shotcrete

Shotcrete is an automated process for manufacturing concrete structures without formwork, using spray concrete. Shotcrete has been used for several years, with skilled operators, as a manual process in the field of mining, tunneling, and other construction [78]. During this process, manual error can influence the quality and finish of the structure. The shotcrete process is now usually fully automated; instead of manual operation, a robotic arm sprays the concrete in a layer-by-layer process [79]. Lindemann et al. [80] have developed a digitally fabricated freeform reinforced concrete structure using an automated shotcrete process. The layer inter-bonding is ensured by the concrete spraying pressure, as well as angle and velocity of spraying. Kloft et al. [81] have investigated the benefits of shotcrete over other DF methods and reported that structural, mechanical properties and geometric accuracy are very high. The thickness, width, angle, and velocity of spraying can be controlled effectively to attain the desired degree of complexity of a structure [81] as shown in Fig.5.

3.1.3 Smart dynamic casting

Smart dynamic casting is unlike other DF methods, and is based on the principle of slip forming rather than printing or spraying [15]. A non-conventional provisional formwork method is used in this system [82]. Several actuators control the shape of the slip forming to cast a structure in the desired shape [83]. The set-on-demand is used in the smart dynamic casting method to sustain its self-weight and to retain shape without collapsing. The growth of slip forming from the early stage to the current automated stage has been summarized by Lloret et al. [84]. Smart dynamic casting is versatile in reinforcement incorporation methods which are similar to concrete 3D printing. The set-on-demand concrete extruded from the pumping system into the dynamic casting formwork. The shape of the slip casting system, its cross-section with horizontal spacing (h₁ to h₆), is dynamically controlled by multi-actuators (A₁–A₆), as illustrated in Fig.6. The automated feedback system [85] controls the movement of the slip formwork with live inspection of material to ensure the movement of slip formwork without distortion or collapse in the prepared structure.

3.1.4 Material intrusion

Material intrusion is also a layer-by-layer printing process, which works based on the principles of selective intrusion of materials using binders or activators, to bind the materials [86]. The material intrusion method can print very complex and organic shapes without any support. The advantages of the material intrusion system have been studied by Lowke et al. [21] and are the independence of shape complexity and printing time, no limitation in overhanging angle, and high surface finish, and reusage of unreacted powder bed particle and limitation of build area. The material intrusion system consists of a feeder, build area, and overflow material collecting chambers as shown in Fig.7. The feed piston pushes the material upward and the recoater transfers the material to the build area; excess material is collected in the overflow chamber. The binder head position is controlled by a gantry or robotic system. After successive layer printing, the build piston moves downwards. The process is repeated for each layer up to completion of the designed geometry. After the completion of the printing process, the unreacted particles and particles in the overflow chamber can be removed and reused.

3.1.5 Mesh molding

Mesh molding is the process of integrating reinforcement as formwork, using sprayed concrete in an automated process. Successive and synchronized fabrication processes are two types of mesh molding [13,14]. The reinforcement is assembled and welded by an automated robot and concrete is sprayed simultaneously by another robotic arm in a synchronized process. This process requires a highly automated system for communication between two robots. In the successive fabrication process, the concrete spraying process starts only after the completion of reinforcement assembly. The successive fabrication process of a double curve wall [14] is shown in Fig.8.

3.1.6 Injection concrete three-dimensional printing

Injection concrete 3D printing is a process of printing or injecting the primary medium (i.e., concrete) in or onto a secondary medium (i.e., colloidal gel) [88]. The secondary medium supports the primary medium in staying in a stable position under gravity. The rheological properties of both media need to be designed optimally to achieve high-quality printing [89]. Injection concrete 3D printing is capable of printing complex and organic shapes. The complexity that can be achieved by this method can be visualized by a complex table printed by Meshkini and Zollner and developed at the Digital Building Fabrication studio in 2019 as shown in Fig.9 [89].

3.1.7 Thin forming

Thin forming [16,17] works based on the principles of a shell and fill method, and it falls into the provisional formwork category. The shell is 3D printed using the FDM process and it is filled with set-on-demand concrete. Hydrostatic pressure is an important factor in thin forming, particularly in the context of the structural integrity of 3D printed formwork, where it influences the risk of bursting. By considering such factors, set-on-demand concrete is used to reduce the hydrostatic pressure and to prevent the deformation or busting of thin formwork. The thin formwork is strengthened using functionally graded carbon-reinforced thermoplastic materials produced by simultaneous and consecutive fabrication methods [90]. In the simultaneous fabrication method [16,17], the set-on-demand concrete is filled simultaneously while the shell is being printed, as shown in Fig.10. However, in the consecutive fabrication method, the concrete fills the shell after it has been completed [91].

3.2 Criteria selection

The selection of criteria is an important aspect of MCDM for ranking and evaluating the alternative techniques. The criteria are formulated based on the 3Ms (i.e., Method, Material, and Machine) of DF as shown in Fig.11. The 3Ms are interlinked and mutually connected with criteria.

3.2.1 Formwork utilization

The cost of the formwork for complex and organic structures is about 60% of the construction cost [91]. Formwork is one of the most time, labor, and cost-consuming components of the construction industry. To achieve minimal material usage in conventional construction practices, formwork needs to be reduced. One of the main objectives of adopting DF in construction is to eliminate formwork. In DF, different approaches to formwork are as follows: 1) no formwork; 2) provisional formwork; 3) incorporated formwork.

Concrete 3D printing and material intrusion methods can print without any formwork. Smart dynamic casting [83,85], injection concrete 3D printing [88,89], and thin forming [16,17] use non-conventional formwork during the printing process alone, hence these processes use provisional formwork. Incorporated formwork is utilized by mesh molding [12], and shotcrete [80,81,92], where the reinforcement acts as the formwork. Based on the literature [43], the criteria can be assessed using a 0−5 scale, as shown in Tab.3. The high values are assigned to concrete 3D printing and material intrusion methods because they can be printed without formworks. On the other hand, provisional formwork and incorporated formwork are categories in a medium range, as shown in Fig.12.

3.2.2 Reinforcement incorporation

Reinforcement incorporation directly influences formwork utilization, geometrical complexity, and assembly complexity. The 3D-printed Zhaozhou Bridge [5], Baoshan Pedestrian Bridge [42], and Striatus arched foot bridge [2] were designed without reinforcement, hence conventional formwork was used during the fabrication process. Moreover, the reinforcement incorporation affected the assembling method of the Nijmegen Bicycle Bridge [41] and Bicycle Bridge [7] as these bridges were printed in segments due to the build area limitations and to reduce transportation complexity. Moreover, post-tensioned reinforcement was used to ensure structural stability. The different types of reinforcement incorporation methods are listed below.

1) Micro reinforcement (MR).

2) Conventional reinforcement (CR).

3) Post-tensioning (PT).

4) Sync reinforcement (SR).

The DF technique may involve multiple reinforcement incorporation methods. The versatility of DF is expressed in a 1–5 scale, as shown in Tab.4, where every reinforcement incorporation method has 1 point except sync reinforcement. The sync reinforcement has 2 points due to automation.

3.2.3 Geometrical complexity

Geometrical complexity is interlinked with the nature of the material used and the reinforcement incorporated. The geometrical complexity that can be achieved with DF is scaled into different categories as listed below.

1) Very High–5.

2) High–4.

3) Medium–3.

4) Low–2.

5) Very Low–1.

Material intrusion [95–100] and injection concrete 3D printing [88,89] have very high structural stability while printing because printing takes place over a material bed or gel, respectively. Thin forming is in the category of shell and fill [16,17], which also has high stability. Concrete 3D printing, shotcrete, and mesh molding fall into the medium stability category. Smart dynamic casting has the lowest potential for complexity because it works on the principle of slip formwork. Some geometrical complexities based on different DF methods are shown in Fig.13. The scaling of potential for geometrical complexity of different alternatives is shown in Tab.5.

3.2.4 Material enhancement

Material properties, including printability, extrudability, and buildability, need to be enhanced in different ways depending on the type of DF technique used. For concrete 3D printing, the material needs to have printability, extrudability, and buildability properties [99,101]. In addition to that, shotcrete and mesh molding need to have sprayability. Set-on-demand [102,103] material is used in the smart dynamic casting process to ensure stability and to hold the shape even after moving the provisional formwork. Among all of the DF techniques, injection concrete 3D printing requires very high material enhancement. Tab.6 shows the material enhancement scaling.

1) Very High enhancement required–5.

2) High–4.

3) Medium–3.

4) Low–2.

5) Very Low–1.

3.2.5 Assembly complexity

The assembly is interlinked with other criteria including build area and reinforcement. The main reasons for segmented printing and assembly onsite are limitations in the build area or to reduce transportation complications. Concrete 3D printing, shotcrete, and mesh molding DF techniques can use a mobile robotic arm to print or a robotic arm with a gantry [73,104]. The assembly of the structure is necessary because of limitations in the build are. Due to the restricted printing space, the structure is printed in segments and later assembled. This issue can be solved by using a mobile robotic arm printer with a larger build area, eliminating the need for post-print assembly. Salet et al. [7] have used conventional casting elements at the end of the 3D-printed bridge to bear the post-tensioning loads. The automatic process of the system can be enhanced by avoiding the conventional casting elements in the assembling process. Meanwhile, Eindhoven University of Technology (TU/e) [41] used the shell and fill method for end bearing in a concrete 3D printed bicycle bridge to overcome problems associated with the usage of conventional casting elements. The material intrusion [95–100] and injection concrete 3D printing [88,89] processes have build area limitations. Tab.7 shows the assembly complexity scaling of DF.

1) No Assembly.

2) Assembly with DF element.

3) Assembly with a conventional casting (CC) element.

3.2.6 Surface finish

The surface finish is an important aesthetic characteristic. A complex surface texture can be produced in thin forming and smart dynamic casting with very high surface quality. Concrete 3D printing has a layer-by-layer finish and a trowel system is used to reduce ripple effect. The DF techniques are scaled according to their surface finish quality as shown in Tab.8.

1) Very High–5.

2) High–4.

3) Medium–3.

4) Low–2.

5) Very Low–1.

3.2.7 Build area

The build area of the printer is one of the deciding factors in fabricating any structure. In DF the build area depends on the printer that is used for the adopted DF technique. The gantry system is a versatile printer that can be modified as per the required size. Residential buildings are printed using concrete 3D printer with a combination of a gantry system. A mobile robotic arm technique [12] does not have any limitations regarding build area and can be adopted in shotcrete and mesh molding. Therefore, these two DF techniques offer desirable build area. One of the main drawbacks of material intrusion and injection concrete 3D printing is build area limitations. These two DF techniques need one medium to print, a powder bed for intrusion, and gel for injection concrete 3D printing. Tab.9 shows the build area scaling.

1) Very High–5.

2) High–4.

3) Medium–3.

4) Low–2.

5) Very Low–1.

A decision matrix is formulated based on the scaling of all criteria as shown in Tab.10.

4 Results and discussions

The subjective weighting method AHP determines the weights for the different criteria. From Tab.11, it can be seen that the order of weight begins with the reinforcement incorporation method (ranked 1), and geometrical complexity (ranked 2), followed by build area (ranked 3), assembly complexity, and material enhancement with the least preference given for surface finish and formwork utilization. On the other hand, the object weighting method ENTROPY gives more preference to reinforcement incorporation (ranked 1) as does AHP. Subsequently, the order continues from build area (ranked 2), geometrical complexity (ranked 3), surface finish, formwork utilization, assembly complexity, with material enhancement as least preferred. From the results, it is observed that the three most preferred criteria (i.e. reinforcement incorporation, geometrical complexity, and build area) are similar in AHP and ENTROPY weighting methods, however, the dissimilarity is greater for the other criteria.

Both AHP and ENTROPY weighting methods are used for three different MCDM techniques, namely, TOPSIS, WASPAS, and MOORA. The overall performance of the alternatives is shown in Tab.12. The ranking-based performance is given for alternatives using three MCDM techniques with two weighting combination methods as shown in Tab.13.

The reinforcement incorporation, build area, and geometrical complexity are the most weighted criteria, which contribute more to the selection process and are the main deciding factors. These three criteria contribute to a nearly 70% of weight scores, as shown in Tab.11.

Concrete 3D printing, shotcrete, and mesh molding perform well in reinforcement incorporation and build area criteria and have average performance in geometric complexity. Subsequently, the smart dynamic casting and thin forming give average performance, having fourth and fifth ranks, respectively. This is due to the low performance in reinforcement incorporation and build area criteria. The material intrusion method secures the sixth rank because of limitations in the reinforcement incorporation method and build area. However, the injection concrete 3D printing technique is at an early stage of development and it needs to improve in reinforcement incorporation methods and build area. Moreover, injection concrete 3D printing is one of DF methods that could produce very complex structures, next to the material intrusion methods.

The combination of three MCDM techniques with two different weighting methods gives a similar ranking for alternatives. From Tab.13, it can be seen that the ranking of smart dynamic casting varies from rank 4 to 5 in AHP-WASPAS.

4.1 Sensitivity analysis

To ensure the sensitivity and integrity of the results, a sensitivity analysis is conducted based on criteria with higher weights. The sensitivity analysis was performed based on the methodology outlined by Raigar et al. [105] and Mangla et al. [106]. This analysis involves evaluating the ranking by systematically adjusting the weights of all criteria in relation to the highly weighted one as shown Tab.14. In this study, the highest weighted criterion is the incorporation of reinforcement. Consequently, its weight is varied within the range of 0.1 to 0.9. Moreover, the weights for the remaining criteria are determined using Eq. (19).

(19)\mathrm{\Delta }j=\frac{\mathrm{\Delta }p\ast w_j}{w_j-1};j=1,2,\dots ,k;j\ne p.

When the weight of criteria P changes by \mathrm{\Delta }p, the weights of the other criteria undergo a corresponding adjustment denoted as \mathrm{\Delta }j. The modification of the weight of the highest-priority criterion, denoted as ‘p’, is determined according to Eq. (20). Furthermore, the weights of the remaining criteria are adjusted according to Eq. (21).

(20)w_p^{\prime }=w_p+\mathrm{\Delta }p,

(21)w_j^{\prime }=w_j+\mathrm{\Delta }j.

Tab.14 illustrates the weights obtained from the sensitivity analysis, whereby the weight of reinforcement incorporation is varied from 0.1 to 0.9. These weight adjustments directly impact the ranking of the alternatives, as depicted in Fig.14.

From Fig.14(a)–14(c), it is apparent that concrete 3D printing consistently maintains the top rank throughout the sensitivity analysis, regardless of the variation in the weight of reinforcement incorporation (ranging from 0.1 to 0.9), across all three MCDM methods. Conversely, injection concrete 3D printing consistently emerges as the least preferred alternative securing the seventh rank in all cases except TOPSIS (0.1). Furthermore, run (0.1) exhibits a comparatively higher discrepancy across all three methods, even though shotcrete and mesh molding consistently secure the second and third ranks in TOPSIS and MOORA, a trend that persists up to the Run (0.4) threshold in both methods. Material intrusion consistently achieves the sixth rank within the range of 0.2 to 0.9 across all methods, but attains the fifth rank in TOPSIS (0.1) and MOORA (0.1). Most notably, thin forming consistently secures the fifth rank, within the range of 0.3 to 0.9, in all three methods, but rises to the fourth rank in TOPSIS (0.1), WASPAS (0.1, 0.2), and MOORA (0.1, 0.2). Upon conducting the sensitivity analysis, it is evident that reinforcement incorporation, build area, and geometrical complexity carry the highest weightage. Concrete 3D printing, shotcrete, and mesh molding perform exceptionally well, consistently securing the top three ranks. Hence results present in this study is consistent and reliable.

5 Conclusions

Selecting the most suitable DF method to mitigate negative environmental impact in the construction industry is a complex endeavor. It is imperative to strike a balance between environmental considerations, cost-effectiveness, and construction efficiency. In this research, a rigorous evaluation was conducted by utilizing three MCDM methods (TOPSIS, WASPAS, and MOORA), integrating both the objective (ENTROPY) and subjective (AHP) approaches. Notably, it was observed that, until now, MCDM approaches have not been widely adopted in the selection of DF methods in the construction industry. The study incorporated the evaluation of seven alternative DF methods: concrete 3D printing, shotcrete, smart dynamic casting, material intrusion, mesh molding, injection concrete 3D printing, and thin forming. Seven essential criteria, namely formwork utilization, reinforcement incorporation, geometrical complexity, material enhancement, Assembly complexity, surface finish, and build area, were employed in the assessment.

1) The application of three distinct MCDM methods, under both objective and subjective weighting, yielded similar rankings, emphasizing the robustness of the methodology.

2) Results confirm that reinforcement incorporation, geometrical complexity, and build area emerged as the most influential criteria exerting a pivotal influence on the process of selection of DF techniques.

3) Concrete 3D printing demonstrated superior performance across all criteria securing the top ranking.

4) The remaining alternatives could be categorized into three groups based on their performance in the highest weighted criteria (reinforcement incorporation, geometrical complexity, and build area), directly determining their respective rankings.

5) Shotcrete and mesh molding exhibited commendable performance by achieving second and third positions, respectively.

6) Smart dynamic casting and thin forming secured fourth and fifth ranks reflecting their moderate performance.

7) Material intrusion and injection concrete printing, due to limitations in reinforcement incorporation methods and build area, occupied the lower ranks among the alternative DF techniques.

8) It is imperative to emphasize the strategic leveraging of the reinforcement incorporation method across all DF techniques, especially in material intrusion and injection concrete 3D printing. These techniques excel in printing highly complex structures without external support, although this very working principle imposes limitations on build area and the re-use of unreacted cement particles from the printing platform.

Moreover, this study opens up a promising avenue for future research endeavors. By enhancing the performance of lower-ranking criteria, such as build area and geometrical complexity, researchers can further optimize the efficacy of emerging DF technologies within the construction domain.

References

[1] YuanP FChenZZhangL. Form finding for 3D printed pedestrian bridges. In: Proceedings of 23rd CAADRIA Conference. Hong Kong, China: CAADRIA, 225–234

[2] ZahaHadidBlockResearch Group (BRG)ETHZurich. Striatus 3D Printed Concrete Bridge—Zaha Hadid Architects. 2021. Available at the website of Zaha Hadid architects

[3] Anteagroup. Our first 3D-printed bridge sees the light of day. 2020. Available at the website of Anteagroup

[4] ZhanQZhouXYuanP F. Digital design and fabrication of a 3D concrete printed prestressed bridge. In: Proceedings of 26th CAADRIA Conference. Hong Kong, China: CAADRIA, 663–672

[5] MaG. The longest- span prefabricated 3D-printed concrete bridge completed at Hebei. 2019. Available at the website of Hebei University of Technology

[6] Pons-Valladares O, Casanovas-Rubio M del M, Armengou J, de la Fuente A. Approach for sustainability assessment for footbridge construction technologies: Application to the first world D-shape 3D-printed fiber-reinforced mortar footbridge in Madrid. Journal of Cleaner Production, 2023, 394: 136369

[7] Salet T A, Ahmed Z Y, Bos F P, Laagland H L. Design of a 3D printed concrete bridge by testing. Virtual and Physical Prototyping, 2018, 13(3): 222–236

[8] Bazli M, Ashrafi H, Rajabipour A, Kutay C. 3D printing for remote housing: Benefits and challenges. Automation in Construction, 2023, 148: 104772

[9] WolfsRBosDSaletT. Lessons learned of project Milestone: The first 3D printed concrete house in the Netherlands. Materials Today: Proceedings, 2023

[10] BurgerJAejmelaeus-LindströmPGürelSNiketićFLloret-FritschiEFlattR JGramazioFKohlerM. Eggshell Pavilion: A reinforced concrete structure fabricated using robotically 3D printed formwork. Construction Robotics, 2023: 213–233

[11] Prasittisopin L, Sakdanaraseth T, Horayangkura V. Design and construction method of a 3D concrete printing self-supporting curvilinear pavilion. Journal of Architectural Engineering, 2021, 27(3): 05021006

[12] Dörfler N, K J, Hack F, N M. Mobile robotic fabrication beyond factory conditions: Case study Mesh Mould wall of the DFAB HOUSE. Construction Robotics, 2019, 3: 53–67

[13] HackNLauerWGramazioFKohlerM. Mesh Mould: robotically fabricated metal meshes as concrete formwork and reinforcement. In: Proceedings of the 11th International Symposium on Ferrocement and 3rd ICTRC International Conference on Textile Reinforced Concrete. Aachen: RILEM Publications, 2015, 1–13

[14] HackN P. Mesh Mould A robotically fabricated structural stay-in-place formwork system. Dissertation for the Doctoral Degree. Zurich: ETH Zurich, 2018

[15] Lloret FritschiE. Smart Dynamic Casting—A digital fabrication method for non- standard concrete structures. Dissertation for the Doctoral Degree. Zurich: ETH Zurich, 2016

[16] BurgerJLloret-FritschiETahaNScottoFDemoulinTMata-FalcónJGramazioFKohlerMFlattR J. Design and fabrication of a non-standard, structural concrete column using eggshell: Ultra-thin, 3D printed formwork. In: Proceedings of Second RILEM International Conference on Concrete and Digital Fabrication: Digital Concrete 2020 2. Eindhoven: Springer International Publishing, 2020: 1104–1115

[17] Burger J, Lloret-Fritschi E, Scotto F, Demoulin T, Gebhard L, Mata-Falcón J, Gramazio F, Kohler M, Flatt R J. Eggshell: Ultra-thin three-dimensional printed formwork for concrete structures. 3D Printing Additive Manufacturing, 2020, 7(2): 49–59

[18] Wangler T, Lloret E, Reiter L, Hack N, Gramazio F, Kohler M, Bernhard M, Dillenburger B, Buchli J, Roussel N. . Digital concrete: Opportunities and challenges. RILEM Technical Letters, 2016, 1: 67–75

[19] Panda B, Tay Y W D, Paul S C, Tan M J. Current challenges and future potential of 3D concrete printing. Materialwissenschaft und Werkstofftechnik, 2018, 49(5): 666–673

[20] Menna C, Mata-falcón J, Bos F P, Vantyghem G, Ferrara L, Asprone D, Salet T, Kaufmann W. Opportunities and challenges for structural engineering of digitally fabricated concrete. Cement and Concrete Research, 2020, 133: 106079

[21] Lowke D, Dini E, Perrot A, Weger D, Gehlen C, Dillenburger B. Particle-bed 3D printing in concrete construction—Possibilities and challenges. Cement and Concrete Research, 2018, 112: 50–65

[22] Bos F, Wolfs R, Ahmed Z, Salet T. Additive manufacturing of concrete in construction: Potentials and challenges of 3D concrete printing. Virtual and Physical Prototyping, 2016, 11(3): 209–225

[23] Salaimanimagudam M P, Jayaprakash J. Optimum selection of reinforcement, assembly, and formwork system for digital fabrication technique in construction industry—A critical review. Structures, 2022, 46: 725–749

[24] Buswell R A, Leal de Silva W R, Jones S Z, Dirrenberger J. 3D printing using concrete extrusion: A roadmap for research. Cement and Concrete Research, 2018, 112: 37–49

[25] de Schutter G, Lesage K, Mechtcherine V, Nerella V N, Habert G, Agusti-Juan I. Vision of 3D printing with concrete—Technical, economic and environmental potentials. Cement and Concrete Research, 2018, 112: 25–36

[26] Kristombu Baduge S, Navaratnam S, Abu-Zidan Y, McCormack T, Nguyen K, Mendis P, Zhang G, Aye L. Improving performance of additive manufactured (3D printed) concrete: A review on material mix design, processing, interlayer bonding, and reinforcing methods. Structures, 2021, 29: 1597–1609

[27] Pan T, Jiang Y, He H, Wang Y, Yin K. Effect of structural build-up on interlayer bond strength of 3D printed cement mortars. Materials, 2021, 14(2): 1–17

[28] Marchment T, Sanjayan J, Xia M. Method of enhancing interlayer bond strength in construction scale 3D printing with mortar by effective bond area amplification. Materials & Design, 2019, 169: 107684

[29] He L, Tan J Z M, Chow W T, Li H, Pan J. Design of novel nozzles for higher interlayer strength of 3D printed cement paste. Additive Manufacturing, 2021, 48: 102452

[30] Lloret-Fritschi E, Scotto F, Gramazio F, Kohler M, Graser K, Wangler T, Reiter L, Flatt R J, Mata-Falcón J. Challenges of real-scale production with smart dynamic casting. RILEM Bookseries, 2019, 19: 299–310

[31] Huang X, Yang W, Song F, Zou J. Study on the mechanical properties of 3D printing concrete layers and the mechanism of influence of printing parameters. Construction and Building Materials, 2022, 335: 127496

[32] He L, Chow W T, Li H. Effects of interlayer notch and shear stress on interlayer strength of 3D printed cement paste. Additive Manufacturing, 2020, 36: 101390

[33] van der Putten J, de Volder M, van den Heede P, Deprez M, Cnudde V, de Schutter G, van Tittelboom K. Transport properties of 3D printed cementitious materials with prolonged time gap between successive layers. Cement and Concrete Research, 2022, 155: 106777

[34] Tay Y W D, Lim J H, Li M, Tan M J. Creating functionally graded concrete materials with varying 3D printing parameters. Virtual and Physical Prototyping, 2022, 17(3): 662–681

[35] IAAC. 3D printed bridge—Spain. 2016. Available at the website of The Institute for Advanced Architecture of Catalonia

[36] Hansemann G, Schmid R, Holzinger C, Tapley J, Kim H H, Sliskovic V, Freytag B, Trummer A, Peters S. Additive fabrication of concrete elements by robots. Fabricate, 2020, 2020: 124–129

[37] KazadiBYaoLWangL. In-process reinforcement method for 3d concrete printing: status, potentials and challenges. In: Proceedings of International Conference on Green Building, Civil Engineering and Smart City. Singapore: Springer Nature Singapore, 2022, 394–402

[38] Tuvayanond W, Prasittisopin L. Design for manufacture and assembly of digital fabrication and additive manufacturing in construction: a review. Buildings, 2023, 13(2): 429

[39] Vantyghem G, de Corte W, Shakour E, Amir O. 3D printing of a post-tensioned concrete girder designed by topology optimization. Automation in Construction, 2020, 112: 103084

[40] JipaAGiacomarraFGieseckeR. 3D-printed formwork for bespoke concrete stairs: From computational design to digital fabrication. In: Proceedings of the 3rd Annual ACM Symposium on Computational Fabrication. Pennsylvania: SIGGRAPH. 2019, 1–12

[41] TU/e. Nijmegen has the longest 3D-printed concrete bicycle bridge in the world. 2021. Available at the website of Eindhoven University of Technology

[42] Burry J, Sabin J, Sheil B, Skavara M, Xu W, Gao Y, Sun C, Wang Z. Fabrication and application of 3D-printed concrete structural components in the Baoshan pedestrian bridge project. Fabricate, 2020, 2020: 140–147

[43] Raja S, Rajan A J. A decision-making model for selection of the suitable FDM machine using fuzzy TOPSIS. Mathematical Problems in Engineering, 2022, 2022: 1–15

[44] Chodha V, Dubey R, Kumar R. Selection of industrial arc welding robot with TOPSIS and Entropy MCDM techniques. Materials Today: Proceedings, 2022, 50: 709–715

[45] Palanisamy M, Pugalendhi A, Ranganathan R. Selection of suitable additive manufacturing machine and materials through best–worst method (BWM). International Journal of Advanced Manufacturing Technology, 2020, 107(5–6): 2345–2362

[46] Agrawal R. Sustainable material selection for additive manufacturing technologies: A critical analysis of rank reversal approach. Journal of Cleaner Production, 2021, 296: 126500

[47] DohaleVAkarteMGuptaS. Additive manufacturing process selection using MCDM. In: Proceedings of Advances in Mechanical Engineering: Select Proceedings of ICAME 2020. Singapore: Springer Singapore, 2020: 601–609

[48] Ren D, Choi J K, Schneider K. A multicriteria decision-making method for additive manufacturing process selection. Rapid Prototyping Journal, 2022, 28(11): 77–91

[49] Yoris-Nobile A I, Lizasoain-Arteaga E, Slebi-Acevedo C J, Blanco-Fernandez E, Alonso-Cañon S, Indacoechea-Vega I, Castro-Fresno D. Life cycle assessment (LCA) and multi-criteria decision-making (MCDM) analysis to determine the performance of 3D printed cement mortars and geopolymers. Journal of Sustainable Cement-Based Materials, 2023, 12(5): 609–626

[50] Chakraborty S. Applications of the MOORA method for decision making in manufacturing environment. International Journal of Advanced Manufacturing Technology, 2011, 54(9–12): 1155–1166

[51] Iranfar S, Karbala M M, Shahsavari M H, Vandeginste V. Prioritization of habitat construction materials on Mars based on multi-criteria decision-making. Journal of Building Engineering, 2023, 66: 105864

[52] MiçPAntmenZ F. A decision-making model based on TOPSIS, WASPAS, and MULTIMOORA methods for university location selection problem. Sage Open, 2021, 11(3): 21582440211040115

[53] Sabaei D, Erkoyuncu J, Roy R. A review of multi-criteria decision making methods for enhanced maintenance delivery. Procedia CIRP, 2015, 37: 30–35

[54] Doke A B, Zolekar R B, Patel H, Das S. Geospatial mapping of groundwater potential zones using multi-criteria decision-making AHP approach in a hardrock basaltic terrain in India. Ecological Indicators, 2021, 127: 107685

[55] Giamalaki M, Tsoutsos T. Sustainable siting of solar power installations in Mediterranean using a GIS/AHP approach. Renewable Energy, 2019, 141: 64–75

[56] Du Y, Zheng Y, Wu G, Tang Y. Decision-making method of heavy-duty machine tool remanufacturing based on AHP-entropy weight and extension theory. Journal of Cleaner Production, 2020, 252: 119607

[57] Du Y W, Gao K. Ecological security evaluation of marine ranching with AHP-entropy-based TOPSIS: A case study of Yantai, China. Marine Policy, 2020, 122: 104223

[58] Zhao D Y, Ma Y Y, Lin H L. Using the entropy and TOPSIS models to evaluate sustainable development of islands: A case in China. Sustainability, 2022, 14(6): 3707

[59] Lin H L, Cho C C, Ma Y Y, Hu Y Q, Yang Z H. Optimization plan for excess warehouse storage in e-commerce-based plant shops: A case study for Chinese plant industrial. Journal of Business Economics and Management, 2019, 20(5): 897–919

[60] HwangC LYoonK. Multiple Attribute Decision Making. Berlin: Springer Berlin Heidelberg, 1981

[61] Zavadskas E K, Turskis Z, Antucheviciene J, Zakarevicius A. Optimization of weighted aggregated sum product assessment. Elektronika ir Elektrotechnika, 2012, 122(6): 3–6

[62] Brauers W K, Zavadskas E K. The MOORA method and its application to privatization in a transition economy. Control and Cybernetics, 2006, 35(2): 445–469

[63] Saaty T L. A Scaling method for priorities in hierarchical structures. Journal of Mathematical Psychology, 1977, 15(3): 234–281

[64] Shannon C E. A mathematical theory of communication. Bell System Technical Journal, 1948, 27(3): 379–423

[65] Gebhard L, Mata-Falcón J, Anton A, Dillenburger B, Kaufmann W. Structural behaviour of 3D printed concrete beams with various reinforcement strategies. Engineering Structures, 2021, 240: 112380

[66] Gebhard L, Burger J, Mata-Falcón J, Lloret Fritschi E, Gramazio F, Kohler M, Kaufmann W. Towards efficient concrete structures with ultra-thin 3D printed formwork: exploring reinforcement strategies and optimisation. Virtual and Physical Prototyping, 2022, 17(3): 599–616

[67] Tay Y W D, Qian Y, Tan M J. Printability region for 3D concrete printing using slump and slump flow test. Composites. Part B, Engineering, 2019, 174: 106968

[68] Roussel N. Rheological requirements for printable concretes. Cement and Concrete Research, 2018, 112: 76–85

[69] Alfani R, Guerrini G L. Rheological test methods for the characterization of extrudable cement-based materials—A review. Materials and Structures, 2005, 38: 239–247

[70] Kaliyavaradhan S K, Ambily P S, Prem P R, Ghodke S B. Test methods for 3D printable concrete. Automation in Construction, 2022, 142: 104529

[71] Khoshnevis B. Automated construction by contour crafting—Related robotics and information technologies. Automation in Construction, 2004, 13(1): 5–19

[72] Asprone D, Auricchio F, Menna C, Mercuri V. 3D printing of reinforced concrete elements: Technology and design approach. Construction and Building Materials, 2018, 165: 218–231

[73] Anton A, Reiter L, Wangler T, Frangez V, Flatt R J, Dillenburger B. A 3D concrete printing prefabrication platform for bespoke columns. Automation in Construction, 2021, 122: 103467

[74] Tho T P, Thinh N T. Using a cable-driven parallel robot with applications in 3D concrete printing. Applied Sciences, 2021, 11(2): 1–24

[75] Mechtcherine V, Nerella V N, Will F, Näther M, Otto J, Krause M. Large-scale digital concrete construction—CONPrint3D concept for on-site, monolithic 3D-printing. Automation in Construction, 2019, 107: 102933

[76] Gosselin C, Duballet R, Roux P, Gaudillière N, Dirrenberger J, Morel P. Large-scale 3D printing of ultra-high performance concrete—A new processing route for architects and builders. Materials & Design, 2016, 100: 102–109

[77] Weger D, Gehlen C, Korte W, Meyer-Brötz F, Scheydt J, Stengel T. Building rethought—3D concrete printing in building practice. Construction Robotics, 2021, 5(3–4): 203–210

[78] Liu G, Zhao J, Zhang Z, Wang C, Xu Q. Mechanical properties and microstructure of shotcrete under high temperature. Applied Sciences, 2021, 11(19): 9043

[79] Heidarnezhad F, Zhang Q. Shotcrete based 3D concrete printing: State of art, challenges, and opportunities. Construction and Building Materials, 2022, 323: 126545

[80] Lindemann H, Gerbers R, Ibrahim S, Dietrich F, Herrmann E, Dröder K, Raatz A, Kloft H. Development of a shotcrete 3D-printing (SC3DP) technology for additive manufacturing of reinforced freeform concrete structures. RILEM Bookseries, 2019, 19: 287–298

[81] Kloft H, Krauss H W, Hack N, Herrmann E, Neudecker S, Varady P A, Lowke D. Influence of process parameters on the interlayer bond strength of concrete elements additive manufactured by Shotcrete 3D Printing (SC3DP). Cement and Concrete Research, 2020, 134: 106078

[82] Lloret-Fritschi E, Wangler T, Gebhard L, Mata-Falcón J, Mantellato S, Scotto F, Burger J, Szabo A, Ruffray N, Reiter L, Boscaro F, Kaufmann W, Kohler M, Gramazio F, Flatt R. From smart dynamic casting to a growing family of digital casting systems. Cement and Concrete Research, 2020, 134: 106071

[83] ScottoFLloret KristensenEGramazioFKohlerMFlattR J. Adaptive control system for smart dynamic casting. In: Proceedings of the 23rd International Conference of the Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong: CAADRIA, 255–264

[84] Lloret E, Shahab A R, Linus M, Flatt R J, Gramazio F, Kohler M, Langenberg S. Complex concrete structures: Merging existing casting techniques with digital fabrication. Computer Aided Design, 2015, 60: 40–49

[85] Lloret FritschiEReiterLWanglerTGramazioFKohlerMFlattR J. Smart dynamic casting slipforming with flexible formwork—Inline measurement and control. In: Proceedings of the 23rd International Conference of the Association for Computer-Aided Architectural Design Research in Asia. Hong Kong: CAADRIA, 12–19

[86] Lowke D, Talke D, Dressler I, Weger D, Gehlen C, Ostertag C, Rael R. Particle bed 3D printing by selective cement activation—Applications, material and process technology. Cement and Concrete Research, 2020, 134: 106077

[87] Hack N, Dörfler K, Walzer A N, Wangler T, Mata-Falcón J, Kumar N, Buchli J, Kaufmann W, Flatt R J, Gramazio F. . Structural stay-in-place formwork for robotic in situ fabrication of non-standard concrete structures: A real scale architectural demonstrator. Automation in Construction, 2020, 115: 103197

[88] Hack N, Dressler I, Brohmann L, Gantner S, Lowke D, Kloft H. Injection 3D concrete printing (I3DCP): Basic principles and case studies. Materials, 2020, 13(5): 1093

[89] Lowke D, Vandenberg A, Pierre A, Thomas A, Kloft H, Hack N. Injection 3D concrete printing in a carrier liquid—Underlying physics and applications to lightweight space frame structures. Cement and Concrete Composites, 2021, 124: 104169

[90] KwonHEichenhoferMKyttasT. Digital composites: Robotic 3D printing of continuous carbon fiber-reinforced plastics for functionally-graded building components. In: Proceedings of Robotic Fabrication in Architecture, Art and Design 2018. Zurich: Springer International Publishing, 2019, 363–376

[91] JipaABernhardMDillenburgerP B. Submillimetre formwork. In: Proceedings of TxA Emerging Design + Technology Conference. Austin, TX: Texas Society of Architects, 2017

[92] Neudecker S, Bruns C, Gerbers R, Heyn J, Dietrich F, Dröder K, Raatz A, Kloft H. A new robotic spray technology for generative manufacturing of complex concrete structures without formwork. Procedia CIRP, 2016, 43: 333–338

[93] Daungwilailuk T, Pheinsusom P, Pansuk W. Uniaxial load testing of large-scale 3D-printed concrete wall and finite-element model analysis. Construction and Building Materials, 2021, 275: 122039

[94] Kloft H, Hack N, Mainka J, Brohmann L, Herrmann E, Ledderose L, Lowke D. Additive fertigung im bauwesen: Erste 3-D-gedruckte und bewehrte Betonbauteile im Shotcrete-3-D-Printing-Verfahren (SC3DP). Bautechnik, 2019, 96(12): 929–938

[95] Utela B, Storti D, Anderson R, Ganter M. A review of process development steps for new material systems in three dimensional printing (3DP). Journal of Manufacturing Processes, 2008, 10(2): 96–104

[96] MarkusK. Solar Sinter. 2022. Available at the website of Kayser Works

[97] Nair S A O O, Alghamdi H, Arora A, Mehdipour I, Sant G, Neithalath N. Linking fresh paste microstructure, rheology and extrusion characteristics of cementitious binders for 3D printing. Journal of the American Ceramic Society, 2019, 102(7): 3951–3964

[98] Kafara M, Kemnitzer J, Westermann H H, Steinhilper R. Influence of binder quantity on dimensional accuracy and resilience in 3D-printing. Procedia Manufacturing, 2018, 21: 638–646

[99] Nair S A, Panda S, Santhanam M, Sant G, Neithalath N. A critical examination of the influence of material characteristics and extruder geometry on 3D printing of cementitious binders. Cement and Concrete Composites, 2020, 112: 103671

[100] Xia M, Nematollahi B, Sanjayan J. Influence of binder saturation level on compressive strength and dimensional accuracy of powder-based 3D printed geopolymer. Materials Science Forum, 2018, 939: 177–183

[101] Arunothayan A R, Nematollahi B, Ranade R, Bong S H, Sanjayan J. Development of 3D-printable ultra-high performance fiber-reinforced concrete for digital construction. Construction and Building Materials, 2020, 257: 119546

[102] Sriramya D N, Ferron R D. Set-on-demand concrete. Cement and Concrete Research, 2014, 57: 13–27

[103] Reiter L, Wangler T, Anton A, Flatt R J. Setting on demand for digital concrete—Principles, measurements, chemistry, validation. Cement and Concrete Research, 2020, 132: 106047

[104] AntonABedarfPYooADillenburgerAReiterBWanglerLFlattTRobertJ. Concrete choreography: Prefabrication of 3D-printed columns. Fabricate 2020: Making Resilient Architecture, 2020: 286–293

[105] Raigar J, Sharma V S, Srivastava S, Chand R, Singh J. A decision support system for the selection of an additive manufacturing process using a new hybrid MCDM technique. Sādhanā, 2020, 45: 1–14

[106] Mangla S K, Kumar P, Barua M K. Risk analysis in green supply chain using fuzzy AHP approach: A case study. Resources, Conservation and Recycling, 2015, 104: 375–390

RIGHTS & PERMISSIONS

Higher Education Press