In the past the global construction industry has developed rapidly in terms of advanced technology and the incorporation of safety and sustainability. Similarly, the Australian construction industry now considers sustainability as a vital factor for maintaining a balance in preservation and consumption of natural resources. To investigate the local trends and general procedures in the selection process for materials in residential building construction, a questionnaire survey was developed and circulated among the leading construction companies in Western Australia.

Survey response rate and validity: Data were collected in three survey formats: postal, online, and self-administered surveys. A total number of 52 survey questionnaires were sent to construction companies with a cover letter and a form for participant consent. From these, 11 responses were received (response rate = 21.15%; Figure 1). This response rate was considered adequate because other researchers in the construction industry also reported response rates for postal surveys between 20% and 30% [1] [2] [3].

For the online surveys, 35 company representatives were sent a uniform resource locator (URL) link to access the survey and record their responses. The response rate was 48.57% (17 responses). Face-to-face, self-administered surveys were also conducted with 7 company representatives (23 companies sent request; 30.43% response rate). This method of data collection is considered the best; complete responses are received through this method. All survey responses were evaluated before analysis to verify that all questions were answered; missing values were adjusted in the analysis with SPSS (Statistical Package for the Social Sciences, (version 24)). Some irrational responses and suggestions were received but most responses were relevant based upon the work experience of the respondents.

We used a Likert scale in our survey, which is considered reliable to investigate the overall perceptions and experiences of a group of relatively homogeneous

individuals of similar backgrounds and trades [4]. Using a multi-criteria decision making (MCDM) technique, we processed the data to develop a sustainability index (SI) for ranking the most appropriate and sustainable building construction materials. The system was validated by running a pilot analysis on local sustainable roofing materials from two major material suppliers.

Margin of error: For an inferential statistical analysis, researchers agree a sample size (n) of more than 30 is required to obtain an accurate value [5] [6] [7] [8]. Prior research validated that minimum threshold value to rank a data acceptable is 30. We got the response from more than 30 construction firms, hence our sample size (n = 36) is acceptable for statistical analysis with the appropriate analysis tool to get acceptable and reliable results.

SPSS (Statistical Package for the Social Sciences, version 24) was used to analyze the data. Frequencies, percentages, averages were calculated. All values are used for open-ended questions. Data were collected on the organization profiles (i.e. type, size, and age of organization; area of building projects, focus, and main client type) to obtain overall snapshots of the organizations.

The implementation of sustainability depends upon the knowledge and awareness of technical stakeholders (e.g. architects, designers, engineers, estimators, and managers).

Although the respondents stated they were aware of sustainability and its importance in construction, 52.77% had only an average knowledge of the sustainable products available in the market. Moreover, the proportion of respondents

Source: Analysis of survey data 2018.

with poor knowledge (22.22%) exceeded the proportion with good knowledge (16.66%) (Table 4). Hence, the major stakeholders need to learn about sustainable products and their efficacy and adaptability. Most respondents thought that clients or their representatives are less concerned about this factor than about other considerations, and so they pay less attention to this issue. The small percentage of respondents with excellent knowledge (8.33%) all belonged to well-established, large organizations.

The gap between information and implementation can be reduced by developing systems and IT features that are easy to use. This is a combined responsibility of all sectors involved in the construction industry. It is a mutual responsibility of government and the private sector to introduce and implement regulations for a “greener” and safer environment for future generations (Raynsford, 2000).

The criteria for sustainable material selection that were investigated were divided into three categories: environmental criteria, socio-economic criteria, and technical criteria (Table 8).

The respondents were asked to rank the sub-criteria for importance and the data were analyzed (Table 9).

Aesthetics was ranked first in the socio-economic category (RI = 0.88; Table 9). This criterion was designated as high importance as per our predefined values: all factors with an RI > 0.8 were considered of high importance and those with an RI < 0.8 considered of medium to high importance.

A decision model will help technical stakeholders to select from a wide range of options, either classical or innovative. We used the above sustainability criteria for computational analysis in order to formulate the best combination of alternative construction materials. Six major criteria were used (Figure 2). These 6 criteria were further divided into 23 sub-criteria (Table 10).

An aggregated MCDM environment means combining the values of a set of attributes to represent a single value for the entire set of attributes. Much work has been done to introduce the prioritization in aggregation method using the TOPSIS. The derivation of our MCDM model using this prioritization approach is described below.

A sustainability index framework basically helps the decision makers to integrate the issues of sustainability while selecting the available construction materials. Selecting sustainable construction materials from the pool of alternative sustainable materials is a time consuming and difficult practice. Applying the MCDM technique is the best method for integrating objective and subjective weights of various conflicting criteria in order to choose the most appropriate sustainable material. However, the process is challenging.

Source: Analysis of survey data (2018).

^aRelative index. ^bmedium to high. ^chigh. Source: Analysis of survey data (2018).

The first step is to determine the most suitable aggregation method, usually from two major types: the crisp aggregation method, which is used to aggregate the real values; and the fuzzy aggregation method, which is used to aggregate the linguistic labels [15].

The second, most complicated, step is to define the boundary condition. The boundary condition compels the result of an aggregation function f(x). The limit is defined as the minimal and maximal boundaries of possible output.

f ( 0 , ⋯ , 0 ) = 0 and f ( 1 , ⋯ , 1 ) = 1 , where x ∈ [ 0 , 1 ] (1)

The commutativity property states that ordering/ranking of arguments does not matter when there is equal importance or no relationship is considered among the different criteria.

f ( x 1 , x 2 , ⋯ , x n ) = f ( x 2 , x 1 , ⋯ , x n ) = f ( x n , x 2 , x 1 , ⋯ ) , x ∈ S (2)

The continuity condition means that the aggregation function does not change markedly’ if small changes were made to the attributes considered in the aggregation process.

U_xÎS[0,1]^x→[0,1] is a continuous aggregation function if:

f ( x ) : [ 0 , 1 ] x → [ 0 , 1 ] (3)

The monotonicity condition implies that aggregation functions are monotonic, that is, the aggregation function shows a “non-decreasing” relationship between the criteria and the aggregation maneuvers.

x ′ i > x i , then f ( x ′ i ) > f ( x i ) where x ∈ S (4)

The idem-potencies condition is an algebraic property that belongs to the binary operation and displays the relationship if:

f ( x , x , ⋯ , x ) = x where x ∈ S (5)

The associativity condition is the ability of the aggregation function to retaliate against the choice of group, which should not influence the overall result of the aggregation process.

f ( f ( x 1 , x 2 , ⋯ ) , x n ) = f ( f ( x 1 , x 2 ) , ⋯ , x n ) , x ∈ S [16] (6)

Many applications require the evaluation of a set of criteria with prioritized relationships within the set in order to reach a conclusion. [17] [18] [19].

The TOPSIS is an approach that originates from the geometric concept of displaced ideal point, which means that the criterion under investigation must be situated between the positive (most favorable) and negative (least favorable) locations [20].

Determining the weights and criteria ranking: Previous studies have discussed many methods for criteria ranking and weight determination [21] [22] [23]. There are two primary methods. The “direct choice of weights” method directly assigns weights based on the opinions and consensus of a group of experts. The “weights determination from data” method derives the weights of criteria from the data available in the same domain for aggregation purposes. We used the first method to get our weighted data set.

Considering the complexity of the data-collection process and the research output, we conducted an empirical inquiry using a real-world, practical scenario. We collected data on roofing materials that are specifically used in residential buildings in Western Australia and are available from the two major suppliers of roofing tiles.

This validation involved the application and evaluation of two suppliers with six tile options. This hypothetical case study was based on the roofing tiles from different suppliers and different options depending on the type of material selected for the test run and the type of structure implemented. Cost was one of the factors considered; however, the most important factors were the six main criteria and related sub-criteria (see Table 10). The details and physical characteristics of the materials are presented in Table 11.

This model analyzes and ranks the sustainable options by using mathematical

implications with the MCDM technique. The data for this pilot study were collected as part of the survey questionnaire. The experts ranked the different alternatives from the suppliers based on the criteria and sub-criteria provided to them. These values were tabulated in an MS Excel^TM spreadsheet and analyzed by running the TOPSIS model. The main factors that drive the selection of roofing materials are:

1) Compliance with Australian building codes and local estate building codes.

2) House type, orientation, and building construction.

3) Local council requirements.

4) Energy management, insulation selection, and the overall anatomy of the building roof.

Roof types have different pitch values depending on building type and house style. The most common pitches for roofing in Australia are 18 - 20 degrees, depending on the materials used. The prime reason for selecting roofing material for the case study is that it can be tailored at the design stage and plays an important role in maintaining the overall sustainability ranking of building.

The three options in Table 11 were tested against six alternatives from the two major suppliers in the Australian construction industry. The alternatives have various competitive advantages over each other.

The tabulated values in the Excel spreadsheet assigned the sub-criteria with positive or negative signs: criteria with an inverse relationship to sustainability are marked as negative and criteria that enhance the sustainability are marked as positive. The six alternatives were tested against the 23 sub-criteria related to those main six. The MCDM technique with TOPSIS gives a final ranking of these criteria and sub-criteria based on expert opinion [24].

Yoon and Hwang introduced the TOPSIS method, which proposes that the best alternative has the shortest distance to the ideal solution [25] [26]. The attribute which favors an alternative material is called the best attribute and the other is called the worst attribute. The goal of this approach is to find the Euclidean space from the ideal solution [27]. TOPSIS comprises six major steps, which are described as follows using our hypothetical problem of roofing material (roof tiles) from the two major suppliers of roof tiling.

Step 1: Calculate the normalized matrix.

The normalization is calculated using Equation (7) [28]:

X i j ¯ = X i j / ∑ j = 1 n X i j 2 where i = 1 , 2 , ⋯ , n and j = 1 , 2 , ⋯ , m (7)

In a matrix, the i and j belong to the first row and first column value.

Step 2: Calculate the weighted normalized matrix.

The normalized matrix is then multiplied with the weighted value as per equation (8);

V i j = X i j ¯ × W j [27] [29] (8)

Step 3: Calculate the ideal best and ideal worst value.

In this step, the ideal best value is the value which suits the criteria, represented by the maximum value; the ideal best negative value which opposes the agreement of criteria is represented by the minimum value.

Step 4: Calculate the Euclidean distance from ideal best value.

This distance will be the closest value to ideal best value, using the Equation (9):

S + i = [ ∑ j = 1 n ( V i j − V J + ) 2 ] 0.5 (9)

Step 5: Calculate the Euclidean distance from ideal worst value.

This distance will be the closest value to ideal worst value, using Equation (10):

S − i = [ ∑ j = 1 n ( V i j − V J − ) 2 ] 0.5 [30] (10)

Step 6: Calculate the relative closeness to the ideal solution.

The relative closeness to the ideal solution is calculated using Equation (11):

C i = S − i / S i + + S i − (11)

The final calculated values rank the alternatives, with 1 being the best alternative (Table 12). The MCDM can be used to solve the problem of material selection where an infinite number of alternatives exist.

Tables 13-17 show the calculations in the Excel spreadsheet. The weighted