Photoreality

A spherical panorama is an image format that represents the whole real surrounding space, covering a 360-degree horizontal angle and a 180-degree vertical angle from the reference point of projection. Generally, spherical panoramas are created with omnidirectional cameras specialized for 360 panorama images or multiple-angle images based on stitching techniques.

Figure 14.4 Construction Hazard Investigation (CHI) and Construction Safety Performance (CSP) modules.

Two main spherical formats are the cubic (6 cube faces (i.e., 6 separate images) to display the whole sphere around the projection point) and the equirectangular (a single 2:1-ratio image covering 360 x 180 degrees) (Arth, Klopschitz, Reitmayr, & Schmalstieg, 2011). To improve the correction of the stitching result, there must be some amount of overlap (e.g., 25%) among pairs of neighbouring images within a unique parallax (Guan, Shark, Hall, & Deng, 2009).

Prototype

Prototype development

Figure 14.5 describes the development of a VP system. Although a spherical panorama is able to present the whole surrounding space, it is not user- friendly in terms of gathering information. Fortunately, the emerging digital projection techniques can completely address this problem. The webGL framework supports smoothly rendering multiple images to form a wide-eye screen. JavaScript code provides more rich features for the user to interact on the screen, such as control functions (e.g. moving around, interchanging scenarios, multimedia resources), supplemental explanation materials (e.g. text, files, links) and management privileges (e.g. system access control, user group) (Ventura & Hollerer, 2013).

On the virtual scenarios, each educational object is marked by a hotspot icon; information appears when the mouse hovers on it. A learner can click to obtain more detailed knowledge (e.g. images, specifications, videos of the

Figure 14.5 VP prototype development.

installation process and additional material). The VR glass view ability will be ready after the external VR glass detection.

Case study

To identify the advantages and limitations of the new learning framework, case studies deriving from real construction accidents are developed for the VP prototype. According to the Occupational Safety and Health Administration (OSHA), the ‘Fatal Four’ (i.e. falls, being struck by an object, electrocution and caught-in cases) are the most common accidents in reality (OSHA, 2019). Thus, the Fatal Four are chosen for designing the 20 VP case studies. They include a fall from a mobile scaffold on the first floor, a worker being struck by a falling object due to the lack of safety nets; electrocution when using hand tool; a worker caught between a truck and concrete due to toppling over of precast concrete building unit, etc.

The educator and students log into the VP system and click CAL functions to learn accident lessons. After thoroughly understanding the ‘Fatal Four’, they continue to experience a virtual high-rise building jobsite in CHI (see Figure 14.4a). The building construction consists of a basement and 14 floors. Students are required to navigate the whole virtual building to investigate potential hazards. Learning activities of students are hazard identification and inspection. Finally, students click the CSP function (see Figure 14.4b) for game-based testing to assess their safety knowledge and skill after using this VP-based learning approach.

Evaluation

Evaluation scheme

To determine the proposed tool’s pedagogic effectiveness and limitations, an evaluation scheme has been designed, depicted in Figure 14.6. The first evaluation stage applies a traditional learning approach: Thirty four-year students participated in a lecture-based lesson led by a construction professor in the classroom. They were asked to complete a before-exam questionnaire.

All students then moved on to the second evaluation stage, which featured adapting the new VP-based learning approach. All students, ten professors and ten safety managers experienced the VP system using their own mobile devices (e.g. ipad, laptop). After the VP experience, all participants evaluated the VP system and new VP-based learning approach through interviews and questionnaires. Likert scales (1 point for strongly disagree to 5 points for strongly agree) were used for these subjective evaluations. The evaluation criteria (Pedro et al., 2015; Usoh, Catena, Arman, & Slater, 2000; Virvou & Katsionis, 2008) for evaluating VP system were 1) Sense of being in real construction workplaces (to what degree users feel they are in a real

Figure 14.6 Evaluation scheme

construction workplace); 2) Ease of navigation (how easily users can navigate the virtual construction jobsite); 3) Comfort of using mobile devices (how comfortable the participant is using the VP system); 4) Close to reality (how well the VP system can provide real-world visibility) and 5) Computer-assisted learning (how well the VP system can support a new learning approach). Tо validate the new learning method, the evaluation criteria (Le et al., 2016; Park, Le, Pedro, & Lim, 2015;) are 1) Learner-oriented approach; 2) Spatial and experiential learning; 3) Learning engagement; 4) Motivation and 5) Improvement of professional skills.

In the final evaluation stage, the thirty students were required to take a final examination after using the VP platform. A comparison of the beforeexam and after-exam results was made to evaluate the learning effectiveness of the new VP-based approach. A paired sample T-test analysis determined whether the difference between the two exam mean scores was statistically significant. A null hypothesis is that the two exam mean scores are equal, while the alternative hypothesis is that two exam mean scores are not equal. SPSS.20 statistic software was used for this paired sample T-test analysis at the 5% significance level.

Result

Figure 14.7 depicts the results of evaluating the VP system, focusing on five aforementioned criteria. All participants responded that they had the feeling of being in a real construction workplace. Moreover, users highlighted that the VP design is intuitive and that they could easily navigate the virtual construction jobsites by using their fingers or a mouse to interact with the system. Since VP system can run based on individual mobile devices, participants felt

Figure 14.7 System evaluation

very comfortable learning anywhere and anytime. They agreed that the proposed system using VP technology provided real-word visibility better than the common 3D-VR. Learners emphasized the importance of computer- assisted learning to overcome the limitations of the traditional approach.

The new learning method evaluation focused on five criteria as illustrated in Figure 14.8. All participants emphasized that the learner-oriented approach prominent in the VP-based method could benefit the learning outcome. Moreover, they emphasized that the VP platform provides good spatial and experiential learning and that the VP-based learning method motivated and engaged them in learning safety' knowledge better than the lecture-based traditional method. Due to their acting as safety managers inspecting hazards and promoting work safely, learners agreed that their safety' professional skills could be improved by obtaining safety' knowledge through the VP system.

Table 14.1 provides average results of both exams, taken by thirty' construction students before and after using the VP system. According to Table 14.2, since the Sig. (2-tailed) value of 0.001 is less than the significance level of

Figure 14.8 Learning method evaluation

Table 14.1 Learning outcome results.

Table 14.2 Paired samples test.

0.05, the null hypothesis was rejected. Therefore, it is concluded that there is a statistically significant difference between the mean scores of the two exams. The evaluation results of learning outcome effectiveness in Table 14.1 reveal that learners using the VP-based method would have higher scores (80.33) than those who leam based on traditional method (75.17). Therefore, it is proven that the proposed VP-based learning approach can assist learners in improving safety knowledge and skills.