Figure 1
The flowchart is divided into three main sections, each enclosed in a dashed box. The sections are interconnected with arrows indicating the process flow. The left section is labeled “Identification of Risk Factors.” It shows a flowchart with four sequence steps from top to bottom: Step 1: Identification of Key Risk Categories and Their Sub-Factors Through Literature Review and Code of Practices. This step has an icon of a yellow warning sign with an exclamation mark and a magnifying glass. Step 2: Contextualization and Validation of Risk Factors of E E O s Through Interviewing Competent Experts in Australia. This step has an icon of two people with a checkmark and speech bubbles. Step 3: Analysing the interviews for identification of new risk factors. This step has an icon of a person analyzing reports. Step 4: Designing a Questionnaire for Data Collection. This step has an icon of a clipboard with a pencil and paper. Arrows connect these boxes downward, indicating the sequence. An arrow from step 4 leads to the second section in the middle. The middle section is labeled “Designing the Fuzzy DEMATEL Survey.” It shows a flowchart with nine sequence steps from top to bottom: Step 1: Questionnaire distribution to the experts. Step 2: Generating the fuzzy initial direct-relation matrix. Step 3: Consistency Rate Calculation. Step 4: Aggregating the direct-relation matrices. Step 5: Normalizing the direct relation matrix. Step 6: Developing the total (direct–indirect) relation matrix. Step 7: Converting the values of the total relation matrix into crisp values. Step 8: Set threshold value and obtain the Impact Relations Map (I R M). Step 9: Generating Impact Relations Map (I R M). An arrow from the “Consistency Rate Calculation” step points rightwards to the next section. The right section has 2 subsections. Both sections are enclosed in a dashed box. The first subsection is labeled “Consistency Rate Calculation.” It shows a flowchart as follows: The flowchart flows vertically from top to bottom, with various shapes and icons connected by arrows. At the top, a rounded rectangle labeled “Start.” Below it, a rectangular box titled “Fuzzy Number Transformation into Vector Format.” This box contains a depiction of multiple stacked sheets or tables with fuzzy numbers transformed into vectors, showing matrices with numerical values such as 0.25, 0.5, 0.75, 1, et cetera. Next, a rectangular process box with a graphic showing overlapping circles containing colored icons representing respondents. The text inside reads, “Calculate the estimation of each respondent with the average of other respondents.” A decision diamond follows with the question, “If smaller collation is less than 0.3, question mark.” The diamond shows two arrows: If “Yes,” it leads to a rectangular process box to the right that says, “Remove the respondent with the lowest solution,” accompanied by an icon of a single respondent figure. The rectangular box leads back to the process box. If “No,” it continues downward. Below the decision box is a process box illustrating “Generate all possible split half,” showing the respondents split into three groups inside overlapping circles. Then, two process boxes follow sequentially: “Compute the correlation value of each split half.” “Compute the average adjusted split-half correlation.” Finally, the flowchart ends with another rounded rectangle labeled “Start,” indicating either a loop or the conclusion of the process. On the left side of the dashed box, there is a large circular icon with arrows in orange and blue, forming a loop around a checkmark. The second subsection is labeled “Sensitivity Analysis.” It also shows a flowchart as follows: At the top, an oval labeled “Start.” Below, a rectangular process box reads, “Initialize the parameters, including N as the number of iterations and W as the weight range.” A small rectangular box with “i equals 1,” indicating the initialization of an iteration counter. Next, a process box with icons of four respondents colored in different colors. The text reads: “Randomly generate the weight of experts in range W.” Below this, a process box reads, “Repeat each response as much as its weight.” Another process box states, “Run DEMATEL,” from which an arrow extends to the “Designing the Fuzzy DEMATEL Survey.” Next, a process box reads, “Determine the region of each factor.” A decision diamond asks, “Is i less than N question mark” with two arrows: If yes, an arrow points right to a small box, “i equals i plus 1,” which then loops back to the “Randomly Generate the Weight of Experts in Range W” box. If no, the flow continues down. Below the decision, a process box reads, “Calculate the probability as the change of region.” At the bottom, an oval labeled “End.” On the left side of the dashed box, there is an icon depicting a stack of graphs and charts connected by lines, representing data analysis or statistics. The process box labeled “Run DEMATEL” leads to the middle section. An arrow from the middle section leads to the second subsection labeled “Sensitivity Analysis.”Research method. Source: Authors’ own work