Next Article in Journal
Examining the Controls on the Spatial Distribution of Landslides Triggered by the 2008 Wenchuan Ms 8.0 Earthquake, China, Using Methods of Spatial Point Pattern Analysis
Previous Article in Journal
Achieving Net-Zero in the Manufacturing Supply Chain through Carbon Capture and LCA: A Comprehensive Framework with BWM-Fuzzy DEMATEL
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Enhancing Soft Skills through Generative AI in Sustainable Fashion Textile Design Education

Department of Fashion Design & Merchandising, College of Social Science, Gachon University, Seongnam-si 13120, Gyeonggi-do, Republic of Korea
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(16), 6973; https://doi.org/10.3390/su16166973
Submission received: 10 June 2024 / Revised: 26 July 2024 / Accepted: 12 August 2024 / Published: 14 August 2024

Abstract

:
This study explores the significance of incorporating soft skill training in fashion design education through the use of artificial intelligence (AI) technology and examines various AI-based approaches for sustainable fashion textile design education employing a multifaceted methodology that encompasses empirical, quantitative, and qualitative methods. We investigate the aspects of Design Sprints, identify key soft skills that help students meet the complex demands of contemporary fashion design workplaces, propose a curriculum guide for AI textile design programs, and evaluate the soft skill training process. Participants included students who had completed basic fashion design courses over three to four semesters and had experience with the fashion design process. The findings confirmed that participants’ soft skills improved across four areas—digital competence, sense of initiative and entrepreneurship, problem-solving and thinking skills, and communication—through the AI-based fashion textile design curriculum. This study validates the importance of integrating AI technology into educational programs to enhance essential soft skills in the digital fashion industry environment. Additionally, it emphasizes the necessity of developing AI technology-specialized design prompts while maintaining a balance between traditional design education and digital design education for sustainable fashion design education.

1. Introduction

The process of developing fashion textile designs has undergone considerable changes over time, driven by the advent of innovative technologies that have introduced new design tools. Recent years have witnessed a growth in interest in using AI in design, which is changing our approach toward textile design. Designers who use AI receive significant attention within the human–computer interaction community, as AI technology aids them by providing deep learning systems that mimic commonly used manual design processes [1,2]. Consequently, AI systems can be effectively incorporated into educational programs for fashion design students.
The process of fashion textile design has also undergone a considerable transformation. Previously, a single designer was responsible for developing a design from start to finish. However, contemporary textile design development often involves collaboration and communication with other designers [3]. Therefore, the ability and willingness to solve problems and think critically with others have become increasingly important. To facilitate team-based design work, Google Ventures has developed project-based teams of small groups of experts called Design Sprints (DSs) [4]. DSs bring experts together for short periods to identify problems and generate new alternatives, demonstrating more efficient and proactive problem-solving than traditional approaches.
Soft skills are the ability to confidently use new technologies, interact with team members, and produce productive project outcomes [5,6]. Significant research has been conducted on the importance of soft skills in the workplace [7,8,9]. These studies suggest that attitudinal skills and dispositions that enable workers to collaborate, communicate, solve problems, and think critically are gaining increasing importance, as these skills are essential for individuals to succeed in modern workplaces. These skills are particularly essential for designers in modern workplaces, where effectively communicating with team members on a project-by-project basis is as important as having creative problem-solving skills. Designers require skills such as ability to communicate, creativity, and ethics to develop designs.
In an era where new technologies such as AI are being used as design tools, a new approach to fashion textile design development is needed [10,11]. Traditional textile design training manuals are no longer adaptable to the current design process. Furthermore, soft skills are becoming an increasingly important competency in various fields, especially after the COVID-19 pandemic.
Currently, fashion textile design curricula in higher education institutions are segmented (from concept development, to design development and refinement, to prototyping) with the aim of acquiring general skills at each stage [12,13,14]. However, in the era of the Fourth Industrial Revolution and in the context of sustainable design, which is functional to advancing the understanding of how the sustainability transition can be realized by transforming products, people’s behaviors, commercial services, cities, and ultimately the entire socio-economic system [15], there are different ways to acquire the necessary skills. Students no longer need to learn formal or mechanical skills to create creative design products. Students, when faced with a problem, can form teams to explore solutions, observe what others have accomplished, implement their own ideas, repeat the process, and achieve their goals. According to Prensky [16], to prepare students for the advanced digital technology-based textile design development environment of the future, it is important that curricula provide hands-on experience in solving various problems that arise in the technology-based design process. Souleles [17] stated that previous teaching approaches were not focused on the needs of learners. For sustainable fashion textile design education, curricula need to be student-centered so that they can cope well with unexpected situations that may arise in practice. Considering the changing design environment and sustainable goals, it is expected that students can efficiently produce technology-based creative output in the workplace if they develop soft skills by learning new textile design methods through AI and by training in the design process of working in teams. If students can learn soft skills at university, they will be able to produce creative output efficiently.
This study explores various approaches to integrating AI technology for sustainable fashion textile design education and discusses the potential and importance of soft skill training through such educational programs. In light of the aforementioned considerations, we have formulated two research objectives:
  • To enhance students’ practical adaptability and competency required in the diversified digital environment of the modern fashion industry, this study aims to identify DSs and soft skills, proposing a curriculum guide for sustainable fashion textile design in AI technology-based design programs at institutions of higher education.
  • Through an AI fashion textile education program based on DSs, this study aims to evaluate the soft skill training process and satisfaction, to ascertain the value and methods of AI utilization as a learning tool to enhance soft skills.

2. Literature Review

2.1. Design Sprint

In its developmental stages, digital product design was influenced by industrial and software designing processes [18]. However, the emergence of digital product design resulted in the creation of a more formal idea-testing framework in the real world. The Google Venture team was the first to conduct the framework development process. A team of experts at Google Venture dedicated themselves for a week to following the process outlined by Google Ventures [4] and progressed rapidly from challenge identification to solution testing in five stages: understanding, sketching, decision-making, prototyping, and testing. Each stage took one day to complete. The team actively designed the concept during the first three days of the sprint. On the fourth day, the expert designers produced a lifelike prototype. Finally, on the fifth day, customers and stakeholders tested the prototype.
DS was created to develop a business model in the engineering field by incorporating a design process to be used within the challenge-based learning (CBL) framework, an evolved form of Problem-Based Learning [19,20,21]. DSs and CBL facilitate the design of high-quality products using User Experience techniques and creativity [16]. Wilkerson and Trellevik [22] confirm the effectiveness of these methods. Furthermore, CBL is an engaging, collaborative approach to education that promotes students’ efforts to learn by solving real-world problems [23]. One main advantage of CBL is its capacity to create practical solutions for complex and genuine issues through application artifact development [24].
According to Souleles [17], past instructional approaches did not prioritize the needs of future learners. The fashion textile design work environment is constantly evolving. In the past, a single designer was responsible for developing a design from start to finish. Currently, however, textile design development is a collaborative effort involving communication with other designers [3]. Attitudinal skills and dispositions, such as critical thinking, collaborative problem-solving, and time-bound idea generation, are becoming increasingly important in the field of design [25]. This team-based project work process has become an essential aspect of fashion textile design, and designers are increasingly adopting this approach.

2.2. What Are Soft Skills?

Soft skills, sometimes referred to as “21st-century skills” or “future work skills 2020” [7,26], are personal transversal competencies such as social aptitudes, language and communication capabilities, friendliness, and the ability to work in a team [27,28,29,30]. Along with other personality traits, these skills characterize interpersonal relationships [23]. They are typically considered complementary to hard skills [24], which are the abilities necessary to perform specific tasks or activities. Success depends on many traits beyond those measured using intelligence quotient, grading systems, and standardized achievement tests [5,6,28]. Soft skills are key competencies that individuals must possess to solve problems and create new products in a flexible society.
Kechagias [8] defines soft skills as the interpersonal (socioemotional) skills facilitating personal development, social participation, and workplace success. Furthermore, according to Ritter et al. [9], employers favor highly educated design graduates who possess soft skills. In a design work environment, soft skills play a crucial role in enhancing graduates’ qualifications. Most studies [24,26,27,28,29,30] on soft skills specify the following abilities required by individuals to adapt to the fourth industrial era: digital competence, a sense of initiative, entrepreneurship, problem-solving ability, and communication ability (Table 1).
The significance of digital competence in the contemporary workplace is increasing rapidly. It involves the confident and critical use of Information Society Technology (IST) in work, organizational culture, and communication [8]. Digital competence involves retrieving, assessing, storing, producing, presenting, and exchanging information, as well as communicating and participating in collaborative networks. Poláková et al. [31] studied the importance of applying soft skills in the workplace, analyzed job postings from 19,000 distinct organizations, and revealed that the most important job requirements were related to soft skills. The results suggest that individuals must balance their proficiencies in soft and digital skills to succeed in a future society characterized by technological advancements. Therefore, the ability to use technology combines the physical and sensory skills required to operate design tools with an understanding of the scientific and technological principles necessary to examine and operate design systems. This ability involves the utilization of computers for searching, evaluating, storing, producing, presenting, and exchanging information, as well as participating and communicating in collaborative networks using the Internet, based on extant knowledge.
Furthermore, a sense of initiative and entrepreneurship refers to the ability to predict the future and apply innovative ideas in real-life situations [33]. This sense is highly valued in today’s business environment as it enables individuals to apply their skills effectively and act appropriately in practical situations. In addition to performing individual theoretical work, students often collaborate on practical tasks to solve workplace challenges. Their sense of initiative helps team workers create business plans and develop new services and products. In the real world, this is an individual’s ability to recognize their work’s context and utilize opportunities, and it requires a specific set of knowledge and experience to ensure job readiness and the performance of industrial activities [26].
Visual communication is a core soft skill, particularly in team-based work environments. Succi and Canovi [34] consider communication skills, commitment to work, and the ability to work in a team to be the most important soft skills that enhance graduate employability. To successfully participate in projects and lead jobs, workers must collaborate not only with their team members but also with relevant individuals [35]. This skill is people-related, similar to interpersonal, teamwork, and customer service skills [8]. Sharing digital experiences visually during team meetings further develops these skills. Moreover, sharing experiences visually within teams motivates individuals to contribute and perform at a high level by exchanging information and expertise. Team-based design ensures the maintenance of expected levels of design quality and the use of appropriate digital technologies in accordance with team goals.
Problem-solving skills enable students to apply their understanding of phenomena to an external context. The ability to purposefully apply problem-solving strategies is essential, regardless of whether the problem and desired solution are clear or require critical thinking and a creative approach to achieve an outcome [8].

2.3. Generative AI-Based Education Programs

In the field of design education, research on design processes using AI is currently underway. Jeon et al. [36] developed AI technology-based creativity support tools to facilitate creative fashion design work. The model externalizes three cognitive operations (extending, constraining, and blending) associated with divergent and convergent thinking based on the traditional fashion design curriculum. Shi et al. [37] determined the creative interaction between designers and AI by reviewing 93 studies, as effective communication between the two actors plays a crucial role in design collaboration. By contrast, the current study revealed the communication barriers between designers and AI technology. As designers have limited input modalities to convey their ideas to AI, proficiency in AI becomes an important factor in generating creative output.
Numerous studies in the field of visualization have utilized AI technology to enhance the expression of ideas and produce optimal results. Zhang et al. [38] used AI to enhance children’s storytelling abilities by creating visual representations of their stories. According to the study, children aged 6–10 years often struggle with expressing their ideas; however, with the help of StoryDrawer, an AI program that supports visual storytelling, they could transform their stories into abstract sketches [38]. Using StoryDrawer, children generated highly creative and elaborate ideas, which resulted in an engaging visual storytelling experience. Huang [39] investigated the development of an intelligent system for visual design and creation utilizing AI technology to assist graphic communication designers in achieving high-quality, efficient, and abundant design output. This study clarifies how to combine AI technology with designers’ design workflow to generate a complementary human–computer mode and provides a packaging design example based on a set of neural network expert systems in various domains.
Recently, attempts have been made to use AI in garment development systems, in both fashion design and manufacturing. Choi et al. [1] developed an automatic fashion design program through human–AI collaboration. The study analyzed existing AI-based garment design tools and developed an AI-based garment development system that incorporates fashion expertise and uses human designers’ garment development process. Lee and Kim [40] investigated whether consumers prefer fashion products designed by AI over those designed by people and found that consumers generally respond more positively to clothes designed by people due to perceived authenticity and expected product quality.
The increasing prevalence of chat GPTs has encouraged researchers to investigate design techniques employing text-to-image generation. Liu et al. [41] suggested a method to construct three-dimensional (3D) models using AI-generated inspiration within Computer-Aided Design software, which was implemented in 3DALL-E, a plugin tool that generates two-dimensional image inspiration for 3D design. The designers who participated in this study highlighted 3DALL-E’s potential within their workflows and used text-to-image AI to produce reference images, prevent design fixation, and inspire design considerations [41].
Although AI-based fashion textile design is an emerging issue, there is currently a lack of substantive discussion on fashion textile design education and design methods that encompass diverse users. To address this gap, it is necessary to develop a design process that considers the range and characteristics of different target users from the perspective of inclusive fashion and can be linked to the design direction.

3. Methodology

The methods were developed to assess the efficacy of the proposed curriculum in enhancing students’ pertinent soft skills. To this end, a multifaceted methodology was employed, encompassing empirical, quantitative, and qualitative methods. Initially, the proposed curriculum was evaluated following the provision of comprehensive details regarding its procedures through the utilization of empirical techniques. Subsequently, quantitative and qualitative approaches were employed, incorporating the utilization of instruments such as the 5-point Likert survey, open-ended surveys, and semi-structured interviews. The research flow for the overall study methodology and design is shown in Figure 1.

3.1. Participants and Project Teams

To collect focus group data, we placed bulletin boards around two campuses in South Korea, announced our program in classes, and recruited students to participate voluntarily. The participants consisted of students who had completed three or more basic fashion design courses over three to four semesters and had a fundamental understanding and experience of the fashion design process. Initially, 35 students volunteered to participate, but only 27 completed the initial and final surveys, yielding a total of 27 responses for this study. Data were collected through an online questionnaire created in Google Drive. All participants were enrolled in a fashion design degree program in South Korea. They were Korean nationals, with 19 women (70.37%) and 8 men (29.63%) aged between 19 and 26 years. All students had completed the traditional fashion design course, which included basic fashion design, fashion illustration, and other related subjects. The training was conducted in five sessions between December 2023 and February 2024. The study protocol was approved by a relevant institutional review board (IRB) in advance.
Before the program commenced, students were required to provide details of their previous experience in fashion design, digital fashion textile design, proficiency in digital design programs, and familiarity with virtual programs, including whether they had utilized 3D CLO. There were variations in students’ experiences in creating fashion textile designs depending on the campus. Thirteen participants (48.14%) in groups 1–4 had no prior experience with digital fashion textile projects, while all fourteen participants (51.85%) in groups 5–9 had previously created digital fashion textiles. The 14 students with prior experience had also used the 3D CLO program, which they used in the prototype process. Thus, all students were initially divided by campus before groups were formed. When forming teams, students were encouraged to present their desired topics to their peers and freely form teams with those who shared similar interests. Allowing students to choose their partners is one way of motivating them to contribute their soft skills effectively [42]. Consequently, students were divided into a total of nine groups, each comprising three to five students (Table 2). Throughout the education program, nine groups were tracked in the fields of AI development and digital avatar application for fashion textiles.

3.2. AI-Based Fashion Textile Design Education Course

In order to address the first research question, which concerns the development of a curriculum guide for fashion textile design, we organized the following curriculum (Table 3). Overall, this study followed the CBL and DS process—Understand–Sketch–Decide–Prototype–Validate. This AI-based fashion textile design program was conducted in person, and an instructor was present in the classroom for every session. Students received feedback on their progress and freely discussed it with their team members. Each session lasted between 60 and 90 min. The objectives and classroom activities for each session are based on the findings of Kwon et al. [43], who studied the process of design inspiration using AI technology, and Ma [44], who developed a fashion design education model by applying the CBL teaching method.
In Stage 1 (Understand), students understand the mechanism of generative AI and are guided through the program to be used in class. The goal of this stage was to enable students to define the problem and set their own objectives. The DS process and project goals were explained to the students; subsequently, they were introduced to design tools utilizing AI. Textiles were developed using Deep Dream Generator and WowPattern, both of which were used in prior studies and considered efficient design tools.
In the second stage (Sketch), students practiced the design using an AI design tool. Each of the nine teams researched design motifs based on the design concept. The teams selected several images and merged them with keywords using the Deep Dream Generator program to create a new image. The resulting image was transformed into a textile pattern using the pattern placement method on the WowPattern website. The AI program refined and extracted the images, and an automatic transformation program rearranged them into textile patterns.
To develop the fashion textile, we adapted an AI-based fashion textile design education course from an earlier study [45] as a textile design development learning model (Figure 2). The course clarified traditional fashion textile development and incorporated AI into technical design issues to conform to a practical curriculum. To integrate the educational model with the existing design curriculum, we organized a five-unit course on fashion textile design utilizing AI.
In the third stage (Decide), students collaborated in teams to enhance the images they had researched to achieve desired outcomes. They practiced AI program manipulation to achieve the desired results and succeeded after undergoing multiple trial and error cycles. Teams convened to share their input and output images and, thereby, learned from each other’s processes and experiences.
In the fourth stage (Prototype), the teams finalized their textile designs for commercialization and applied them to a virtual clothing program by adding the textile pattern to a virtual avatar. All groups applied their textile designs to playable avatars in the program called Zepeto, which was user-friendly even without prior experience. Groups that could utilize 3D CLO, an industrial practice software, used Zepeto (mobile version) and 3D CLO (version 7.2) program.
In the final stage (Validate), the teams presented their designs and received feedback from other teams on effectively presenting their concepts and apparel products. They examined various issues and shared their decisions with the other team members. Finally, participants completed reviews and surveys to provide their opinions on the entire program.

3.3. Soft Skill Evaluation

Once the program concluded, the students completed a survey, and their final designs were evaluated by their peers. The survey was administered using Google Docs to ensure student anonymity and consisted of four sections: digital competence, sense of initiative and entrepreneurship, problem-solving and critical thinking skills, and communication. These sections corresponded to the elements of soft skills described by Murni et al. [26] and Arce et al. [19]. A Likert scale was employed to rate each subskill in the questionnaire, with each aspect comprising two to seven subskills. The questionnaire had an open-ended format to gather opinions on each skill. Table 4 describes the four elements in detail. Following the completion of all assessments, one student from each group was randomly selected for a 30 min online follow-up interview. During the interview, participants were asked to complete a semi-structured questionnaire to justify their ratings.

3.3.1. Digital Competence

Torres-Coronas and Vidal-Blasco [46] define digital competence as the ability to understand and express information through the analytical, productive, and creative use of information technologies and social software to transform information into knowledge. According to this definition, students require digital competence for informed decision-making and collaborative learning with the help of digital tools. To assess the development of this competency, the sub-questionnaire, based on Conlon and Gallery’s study [47], included design expression, efficiency, and work productivity. Through stages encompassing understanding, sketching, decision-making, and prototyping, students acquired digital competencies in IST.

3.3.2. Sense of Initiative and Entrepreneurship

This skill refers to an individual’s ability to translate ideas into actions and tackle challenges. It encompasses the ability to use AI to achieve goals and to plan and manage projects. Based on a study conducted by Conlon and Gallery [47], participants were asked to evaluate the effectiveness of their training program in terms of conceptualizing designs, problem-solving, and product development.

3.3.3. Visual Communication

This section examines how teamwork helps accomplish tasks using AI’s image-generating technology, with an emphasis on improving collaboration and communication skills. The sub-evaluations were structured based on Tran’s [48] framework. Within this subsection, students were surveyed on the effectiveness of AI technology in facilitating visual communication, sharing topics with teammates to enhance understanding, and evaluating the program’s ability to aid in sharing their processes and results with other teams.

3.3.4. Problem-Solving

During the program, students employed AI techniques to tackle their assigned challenges. This survey on problem-solving is based on Tran’s [48] research. Students assessed their ability to effectively apply digital strategies to solve problems, a task that requires critical thinking and a creative approach to achieve results in situations where the problem and desired solution are clear.

4. Results

4.1. Design Work Results

The design processes and outcomes of all students participating in this program indicate that very specific and detailed images can be generated through automatic motif generation based on AI. In the prototype stage, outcomes varied depending on the use of 3D CLO. Groups 1–4, using only Zepeto, are illustrated in Figure 3, whereas groups 5–9, employing 3D CLO and Zepeto, are depicted in Figure 4. Groups 1–4 applied AI-generated textile images to avatars in Zepeto to simulate fashion items and styles, proposing initial design concepts. Conversely, groups 5–9 utilized 3D CLO to demonstrate design elements and new pattern configurations, suggesting simulations that could lead to prototypes.
Significant differences were observed in the design processes from inspirational images and textiles to final prototypes. In the initial sourcing stage, groups 1–4 entered simple keywords such as “ballerina”, whereas the more digital-experienced groups 5–9 inputted more complex phrases and images, such as “reef in watercolor”, providing clearer conceptual images.
Furthermore, during the automatic pattern generation stage, differences in preferred arrangement styles of textile design images were noted. Among groups 1–4, 10 out of 13 preferred simple arrangements, while the majority of groups 5–9 favored complex arrangements like facade and cross brick. However, the initial textile arrangement results did not exhibit significant differences, likely due to the limited arrangement options of WowPattern, a motif arrangement program.
Finally, in the digital textile application stage, groups 5–9 modified and manipulated the AI-generated textile designs to align with their design concepts rather than merely matching the items. This suggests that prior digital design experience influenced the design process utilizing AI.

4.2. Evaluation of the Fashion Textile Design Process Using AI

To assess the effectiveness of the fashion textile design curriculum utilizing AI, addressing the second research question, students were asked to rate the entire curriculum on a 5-point Likert scale ranging from 1 = not important to 5 = very important. We analyzed the quantitative data to ascertain whether the integration of AI improved students’ soft skills in the fashion textile design process. In terms of their satisfaction with the program, students awarded an average rating of 4.44 (Table 5), indicating a high level of satisfaction. The standard deviation (SV) is relatively low at 0.64, indicating that there is relatively good agreement among respondents. Ultimately, our proposed curriculum was found to meet students’ expectations, to determine the value and methods of AI utilization as a learning tool, ultimately enhancing soft skills. Additionally, insights from the essay questionnaire data revealed that groups 1–4 expressed a desire for more training in digital tools such as Adobe Photoshop and suggested conducting AI design training earlier in the curriculum to better comprehend the development process. However, given their experience in fashion design using digital technologies, groups 5–9 did not comment on the need for additional technical training and appeared highly content with the current program.

4.3. Result of Digital Competence

In this section, students were asked to evaluate the design expression, efficiency, and work productivity of AI in fashion textile design work (Table 6). The average score for each question exceeded 4 out of 5, indicating that the majority of respondents perceived the integration of AI as a valuable addition to their digital design skills. It is evident that AI has positively impacted fashion textile design. Regarding the digital competence aspect of soft skills, students rated AI technology highest for speed of work, with a ranking of 4.62 at most, indicating that their utilization of AI technology in the fashion textile design process enabled them to work faster. The SV is relatively low at 0.63, which is evidence that all participants have the same opinion. When questioned about whether AI technology would replace traditional digital design tools (question 7 in Table 6), all students indicated their belief that this would not occur. However, the SV is a bit higher at 1.0, which is an indication of a bit more variation in responses among respondents. This response is further explored in the subsequent section of the questionnaire, concerning the design work process. In terms of their satisfaction with productivity, ease of work, and design quality, participants provided ratings of 4.36, 4.54, and 4.21, respectively. The standard deviations show that participants scored relatively close to the mean with 0.74, 0.64, and 0.80. This suggests that participants anticipate AI will support their design work and enhance the efficiency of the fashion textile design process.
When we asked students to provide feedback on AI technology through open-ended questions (Table 7), they expressed satisfaction with its speed and ease of use, which aligns with the results obtained from the Likert scale evaluation. AI technology provided an effective means of expressing their design intentions, a time-saving tool, and a facilitator of their work, ultimately enhancing the quality of their designs. It also provided avenues for increasing work productivity by offering more design options. However, students noted disappointment with the need for trial and error to achieve expected image results and the inability to fine-tune image size and quality. To further enhance the digital competence of soft skills through AI in the future, it is likely that technological or educational advancements will be required to overcome these challenges.

4.4. Result of Sense of Initiative and Entrepreneurship

In this section, students were asked to evaluate how their skills and design work process were enhanced by employing AI technology in fashion textile design. The results revealed that all students responded positively, giving scores of 4 or higher for most questions. Students exhibited a positive response to whether they were inspired by AI, with an average score of 4.29. Similarly, they gave a high score of 4.37 when asked if they were helped by AI fashion textiles. However, when compared to other traditional digital tools, they gave it a relatively low score of 4.13. This suggests that while the design process is satisfactory when working with AI technology, it may not be perceived as superior to traditional design tools.
The final section of the survey, which was open-ended, provided insights into the reasons behind this phenomenon. In Table 8, questions 4, 5, and 6 examined whether creating a fashion product would benefit students. The average response to this question was above 4.3. The mean score for each question was greater than 4 out of 5, indicating that respondents believed that the use of AI in fashion design would facilitate the development of their own designs, assist them in creating fashion products, and be economically viable for fashion product development. Furthermore, respondents indicated that the future fashion industry would require digital fashion textile design skills using AI.
However, when asked whether they anticipated replacing a fashion designer’s work, students gave a rating of 3.77 for question 7. The SV is highest at 1.09, which is an indication of a wide range of opinions among respondents. The lowest mean and highest SV score were recorded for this question, prompting us to conduct additional interviews. This implies that while AI can assist in developing products that are more efficacious in their functionality, students are skeptical about its potential to entirely replace the work of a fashion designer. The students provided the following explanation for their low scores: “AI technology, which is solely data-driven, can produce a result to a certain extent, but it fails to capture the human emotions and experiences involved in designing.” This suggests that even in the era of AI-based design, professional designers still need to possess emotional and experiential skills. Therefore, while AI can be a valuable tool in fashion textile design, it is constrained in its capacity to generate unique designs based on the designer’s emotions and experiences. The designer’s work methodology, encompassing technical and design skills, remains vital to the effective utilization of AI.
Furthermore, participants were required to evaluate the advantages and disadvantages of the AI design process (Table 9). One of the negative aspects identified was the lack of precision, particularly in terms of positioning and resizing. Participants expressed that while the process helped them quickly develop new designs, it could not guarantee creative results. This raises the possibility that some designers may not yet fully trust AI, or that AI may not be appropriate for all design situations. This can be interpreted as a factor in the lower satisfaction with AI design methods compared to design work with traditional digital tools, as identified in question 3. While AI offers a faster pace of work, it does not guarantee the quality of the design work.

4.5. Result of Visual Communication

Table 10 reveals that students scored very high, 4.29, for visual communication, indicating that AI technology improved their communication skills. Furthermore, the students reported that feedback from other teams helped them improve their design work and enhance intra-team communication. The mean score for the question regarding visual communication for design topics (number one in Table 10) was 4.35, indicating that effective communication with team members is crucial for the success of designing fashion textiles using AI. The mean score for the question regarding the utilization of AI in the design process (number two in Table 10) was 4.29, suggesting that the AI training course facilitated the sharing of the design process with team members. The mean score for the question inquiring whether the utilization of AI facilitated feedback from other teams (#3 in Table 10) is 4.36, indicating that the integration of AI into the design process is crucial for enhancing the quality of the final product. These findings demonstrate that the application of AI-powered design processes facilitates the communication of visual information, thereby enhancing the effectiveness of collaborative endeavors.
These findings revealed that students should have sufficient experience with digital technology to effectively communicate with others using AI technology. Furthermore, an in-depth understanding of digital tools is necessary to ensure effective communication and a clear understanding of desired outcomes. This suggests that incorporating digital literacy training alongside AI-based design education can further enhance students’ ability to communicate and collaborate effectively in the context of fashion textile design.

4.6. Result of Problem-Solving

Participants were asked to evaluate their problem-solving abilities after completing the AI-based fashion textile design education course (Table 11). Overall, the students rated the course to be helpful for their development and practical application (4.32) and reported an improvement in their design process (4.11). All participants had similar views on improving the current fashion textile design process. These results indicate that the integration of AI technology into the design education curriculum effectively enhances students’ problem-solving skills and contributes to their overall improvement in the design process.
The results of this survey indicate that the course was beneficial in the development and application of fashion textile design (question #1 in Table 11). The analysis of the course also suggests that the expectation is that training with AI will improve the existing fashion textile design course (question #2 in Table 11).
The problem-solving component entailed the creation and evaluation of a physical prototype in situations where the problem and desired solution were clearly defined. The simulated fashion textiles created by Zepeto and 3D CLO can enhance students’ problem-solving skills in various ways. When students were asked to assess the appropriateness of purposeful design for AI technology through a multiple-choice question (Figure 5), they responded that AI technology is better suited for creating promotional effects, such as virtual avatars, marketing, and advertising, rather than physical prototypes. This aligns with Choi’s [49] findings on the use of 3D digital technology, where fashion designers reported that digital technology was more effective for promotion rather than the purchase of fashion textiles. These results suggest that while AI technology can facilitate problem-solving in certain aspects of fashion textile design, its strengths lie primarily in enhancing promotional aspects rather than physical prototyping.
Participants were asked open-ended questions regarding the challenges faced during fashion textile development and the manner in which they overcame the problems (Table 12). All students reported difficulties in selecting appropriate keywords and images and mentioned the need to provide multiple inputs to overcome these challenges. By inputting keywords into an AI system, we confirmed that workspace search results were often evaluated as unexpected, particularly for new searches; this can be problematic. If AI design classes are to be included in the formal curriculum, students should be provided with a manual explanation of the results generated according to input values. This enables them to obtain more efficient and detailed design outcomes. Some students did not perceive any new or creative aspects of AI and experienced difficulty adjusting the image size and did not notice any significant differences from the existing design.

5. Discussion

5.1. Enhancing Soft Skills in Digital Fashion Industry Environment

To address whether the use of AI improved students’ soft skills in the fashion textile design process, as queried in second research question, we found that participants were satisfied with the fashion textile design course utilizing AI, reporting improvements in their soft skills. The participants expressed satisfaction with the course, with digital competency and communication receiving the highest average scores (4.35 and 4.34, respectively), and the highest number of participants reporting skill enhancement. Additionally, initiative, entrepreneurship, and problem-solving (4.2 and 4.27, respectively) were noted as improved soft skills. The research findings presented herein suggest significant enhancements for fashion textile design development, but also highlight the necessity for understanding and training in AI prompts and virtual prototype programs.
According to Deming [7] and Murni et al. [26], soft skills are pivotal in the fashion design industry, and AI technology is anticipated to aid students in developing these skills for future employability. This study underscores high expectations regarding the speed and ease of use of AI technology and provides comprehensive insights. Consequently, institutions of higher education should consider offering AI technology courses to augment students’ soft skills and employability. As students are instructed in AI technology for fashion textile design, they may become closely associated with a specific piece of new technology. Therefore, it is crucial for them to recognize the existence or development of a wide range of digital technology environments [50].
Sun and Zhao [51] propose that designers and makers in the digital fashion industry are anticipated to establish a more intimate relationship than previously observed. Designers’ roles are expanding and becoming increasingly complex, while makers are expected to evolve into educators collaborating with designers from various fields. Future users will likely focus on reconfiguring traditional manufacturing flows for enhanced efficiency and flexibility. Recognizing the significance of understanding this broader context for their education and future careers in the field, soft skills can aid fashion industry professionals in initiating effective problem-solving and communication with others.

5.2. Development and Implementation of an AI-Based Fashion Textile Design Curriculum

This study identified two critical considerations for designing a curriculum incorporating AI technology: First, it is advisable to integrate AI technology into a semester-long curriculum preceded by a foundational course on existing digital design technology or structured as a semester-long program. While the current program spans five design sessions, it is anticipated that a 15-week, one-semester class, coupled with the development of fundamental digital skills, would significantly enhance students’ understanding of the digital textile development process. According to Kwon et al. [39], digital experience influences design satisfaction, particularly when professional designers evaluate the creativity of AI technology compared to that of students. Moreover, AI digital technology alone is insufficient to produce a finished design product; hence, it is recommended to structure a curriculum adaptable to various types of design classes. By integrating these technologies into practical, entrepreneurial, and digital design classes, students’ soft skills are likely to improve, thereby enhancing their confidence in employment and performance.
Second, students utilizing AI technology should receive prior instruction on selecting appropriate images and keywords. Participants in the current study reported achieving their desired outcomes through manipulation of the technology. The primary limitation of current AI-based tools for garment design development is their inability to accurately reflect designers’ intentions [1]. However, providing students with advance guidance enables them to execute projects more efficiently. Recently, chat GPT technology has enabled designers to work with text-to-image capabilities. Therefore, educating students on textile prompts in advance is expected to deepen their understanding of AI technology and soft skills. Additionally, future studies should utilize updated and recently developed programs to apply AI technology, ensuring that students achieve desired research outcomes.

5.3. Enhancing the Value of Integrating AI Technology in Sustainable Design Education

Students in this program were instructed on AI techniques and tasked with prototype production, emphasizing the refinement of their ideas over time. Participants demonstrated an enhanced understanding of the potential and limitations of generative AI tools and acquired skills in manipulating subject matter for more effective results. The iterative process encouraged students to delve into and experiment with their creative ideas, fostering a deeper appreciation of the possibilities offered by AI tools. While acknowledging ethical concerns regarding copyright and the potential displacement of artists, students recognized the value of generative AI tools in enriching their sketchbooks and ideation process. The project involving AI-generated fashion textiles proved conceptually intriguing, prompting further exploration and consideration within other foundational design classes to develop their ideas fully. To offer insights and value to future AI users, it is imperative that basic design education accompanies the effective utilization of AI. Such a curriculum should incorporate fundamental design lectures alongside AI technology training [52]. This combined approach would also positively impact the development of prompts tailored to each project’s objectives. Moreover, groups 5–9, with greater access to digital design programs, utilized AI technology to generate a diverse array of design outcomes. Hence, it is recommended for sustainable design education that future designer education curricula provide a holistic introduction to both traditional design education and a variety of digital design programs.
Although participants acknowledged the advantages of AI technology in fashion textile design, they expressed skepticism when asked if AI could replace designers. This sentiment underscores the multifaceted nature of designer roles, which require a broad skill set beyond mere design proficiency. Subsequent interviews revealed that designers cannot be wholly substituted by AI technology in certain design realms. Many students emphasized that human touch and experience, elements not replicable with data, are indispensable in these areas, which are not directly linked to soft skills The findings of the participant interviews were consistent with this perspective. A participant posited that while AI is capable of producing results based on data, it is unable to replicate the distinctive emotional experiences and subjective realities of humans:
There will always be tasks that only humans can perform. Cultural nuances, for example, are uniquely human aspects that require a specific background to fully understand and execute. While AI can process vast amounts of data, it can only make decisions based on the information it has been given. Humans, on the other hand, have accumulated knowledge and experience over many years, which allows them to make informed decisions beyond the scope of AI. There will undoubtedly be a distinction between the design resulting from such experiences and the design generated by the computer.
(A participant)
Future sustainable design education should prioritize adapting the curriculum to place greater emphasis on the conceptual framework of creativity. Additionally, establishing a course to educate students on the appropriate utilization of text prompts for AI-generated designs could deepen their comprehension and anticipation of diverse ideation processes [53]. It can be argued that the ability to manipulate algorithms will become a future domain of expertise for artists, akin to computer scientists. The integration of AI into artistic practices and creative workflows will become increasingly common, with prompt engineering assuming a progressively significant role in the process. Sustainable design education must balance the integration of these technologies with traditional design education, providing students with a comprehensive skill set to succeed in an evolving design environment.

6. Conclusions and Limitations

In the ever-evolving landscape of design work, characterized by ambiguous problems and unforeseeable crises, DSs emerge as a strategy enabling teams of experts to tackle issues in focused bursts of time. Through a DS-based sustainable fashion textile design training program utilizing AI, this study delved into how AI technology can enhance soft skills critical for effective problem-solving in team-based fashion industry work. Participants universally reported AI technology’s positive impact on their soft skills across four key areas: digital competence, sense of initiative and entrepreneurship, problem-solving and thinking skills, and communication.
This study underscores the significance of soft skills in the fashion design industry and posits that AI technology can facilitate students’ acquisition of these skills, thereby enhancing their employability. This study underscores the necessity for higher education institutions to provide instruction in the field of artificial intelligence. As students engage with AI for fashion textile design, it is essential for them to possess an understanding of the diverse digital technology environments in which this technology is utilized.
Sun and Zhao [51] posit that designers and makers in the digital fashion industry will form closer relationships, with makers evolving into educators. This transition will necessitate the reconfiguration of conventional manufacturing processes to enhance efficiency and flexibility. Soft skills will be pivotal for effective problem-solving and communication.
The iterative process inherent to AI tool usage encourages students to explore creative ideas, despite ethical concerns such as copyright and artist displacement. AI tools enrich students’ sketchbooks and ideation processes. This study suggests integrating fundamental design education with AI technology training to provide valuable insights to prospective AI users.
This study has two main limitations. First, this study was limited by the educational environment to use a free generative AI program. Despite validation in a previous study [35], future research should employ the latest programs and tailor prompts specifically for fashion textile design. As AI technology rapidly evolves, studies are expected to apply more up-to-date programs and provide appropriate prompts for generative AI focused on fashion textile design. As AI technology is constantly evolving, studies using the latest programs will yield more effective results. Secondly, the small sample size limits the generalizability of the findings. Participants were recruited from a practical fashion design workshop, which may not be representative of the larger population. Although the number of participants was small, more in-depth content was elicited through in-depth interviews.
This research underscores the necessity for comprehensive education on AI technology and its limitations. It advocates for educating students on selecting suitable images and keywords for AI manipulation to efficiently achieve desired outcomes. While recognizing AI’s inability to wholly supplant human designers, students valued AI tools for enhancing their sustainable processes. This study proposes sustainable future design education curricula integrating both traditional and digital design education to furnish students with a comprehensive skill set. Moreover, it underscores the importance of emphasizing the conceptual framework of creativity and educating students on the judicious utilization of AI-generated designs.

Author Contributions

Both authors developed the research idea, analyzed the data, and prepared the manuscript. S.S. guided the overall process of the research and revised the manuscript. D.J. was mainly responsible for data collection and analysis, along with writing the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Gachon University research fund of 2023 (GCU-202304740001).

Institutional Review Board Statement

This research was conducted with the approval and under the supervision of Gachon University Institutional Review Board (IRB Approval No.: 1044396-202310-HR-216-01) regarding ethical issues including consent to participate.

Informed Consent Statement

Informed consent was obtained from all subjects involved in this study.

Data Availability Statement

The data utilized in this research are not publicly accessible owing to continuing follow-up analysis. Nonetheless, it can be accessed through the corresponding author upon a reasonable appeal and with the author’s institutional approval.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Choi, W.; Jang, S.; Kim, H.Y.; Lee, Y.; Lee, S.G.; Lee, H.; Park, S. Developing an AI-based automated fashion design system: Reflecting the work process of fashion designers. Fash. Text. 2023, 10, 39. [Google Scholar] [CrossRef]
  2. Modran, H.A.; Ursuțiu, D.; Samoilă, C. Using the theoretical-experiential binomial for educating AI-literate students. Sustainability 2024, 16, 4068. [Google Scholar] [CrossRef]
  3. Kim, S. Effect of action learning on digital storytelling education for fashion students. Int. J. Fash. Des. Technol. Educ. 2022, 15, 331–341. [Google Scholar] [CrossRef]
  4. Google Ventures. The Design Sprint. 2019. Available online: https://design-sprint.com/google-ventures-design-sprint/ (accessed on 2 April 2024).
  5. Crosbie, R. Learning the soft skills of leadership. Ind. Comm. Train. 2005, 37, 45–51. [Google Scholar] [CrossRef]
  6. Heckman, J.J.; Kautz, T. Hard evidence on soft skills. Lab. Econ. 2012, 19, 451–464. [Google Scholar] [CrossRef] [PubMed]
  7. Deming, D.J. The growing importance of social skills in the labor market. Q. J. Econ. 2017, 132, 1593–1640. [Google Scholar] [CrossRef]
  8. Kechagias, K.; Welsh, M.; Stewart, M.; Mearns, A.; Papadopoulou, D.; Agapidou, E.; Kalivas, I.; Ananiadis, P.; Wåglund, P.; Jonsson, E.M.; et al. Teaching and Assessing Soft Skills, Mass Project; Second Chance School of Thessaloniki: Neapolis, Greece; Thessaloniki, Greece, 2011; ISBN 978-960-9600-00-2. [Google Scholar]
  9. Ritter, B.A.; Small, E.E.; Mortimer, J.W.; Doll, J.L. Designing management curriculum for workplace readiness: Developing students’ soft skills. J. Manag. Educ. 2018, 42, 80–103. [Google Scholar] [CrossRef]
  10. Bozkurt, A.; Karadeniz, A.; Baneres, D.; Guerrero-Roldán, A.E.; Rodríguez, M.E. Artificial intelligence and reflections from educational landscape: A review of AI Studies in half a century. Sustainability 2021, 13, 800. [Google Scholar] [CrossRef]
  11. Murray-Rust, D.; Lupetti, M.L.; Nicenboim, I.; Hoog, W.V.D. Grasping AI: Experiential exercises for designers. AI Soc. 2023, 1–21. [Google Scholar] [CrossRef]
  12. Khurana, K. The Indian fashion and textile sector in and post COVID-19 times. Fash. Text. 2022, 9, 15. [Google Scholar] [CrossRef]
  13. Richardson, T. The global scapes of postmodernity: A proposed model for ‘global cultural flow’ in fashion education. Fash. Theor. 2021, 25, 819–835. [Google Scholar] [CrossRef]
  14. Calamari, S.; Hyllegard, K.H. The process of designing interior textile products & the influence of Design for the Environment (DfE). Fash. Text. 2015, 2, 1–17. [Google Scholar]
  15. Gaziulosoy, I.; Erdoğan Öztekin, E. Design for sustainability transitions: Origins, attitudes and future directions. Sustainability 2019, 11, 3601. [Google Scholar] [CrossRef]
  16. Prensky, M. EMPOWERED!: Reframing “Growing Up” for a New Age; EAI Press: Gent, Belgium, 2022; ISBN 978-0578261409. [Google Scholar]
  17. Souleles, N. Design for social change and design education: Social challenges versus teacher-centred pedagogies. Des. J. 2017, 20 (Suppl. 1), S927–S936. [Google Scholar] [CrossRef]
  18. Banfield, R.; Lombardo, C.T.; Wax, T. Design Sprint: A Practical Guidebook for Building Great Digital Products; O’Reilly Media. Inc.: Newton, MA, USA, 2015; ISBN 978-1491923177. [Google Scholar]
  19. Arce, E.; Suárez-García, A.; López-Vázquez, J.A.; Fernández-Ibáñez, M.I. Design Sprint: Enhancing STEAM and engineering education through agile prototyping and testing ideas. Think. Ski. Creat. 2022, 44, 101039. [Google Scholar] [CrossRef]
  20. López-Caudana, E.; Ruiz, S.; Calixto, A.; Nájera, B.; Castro, D.; Romero, D.; Luna, J.; Vargas, V.; Legorreta, I.; Lara-Prieto, V.; et al. A personalized assistance system for the location and efficient evacuation in case of emergency: TECuidamos, a challenge-based learning derived project designed to save lives. Sustainability 2022, 14, 4931. [Google Scholar] [CrossRef]
  21. Sukackė, V.; Guerra, A.O.P.C.; Ellinger, D.; Carlos, V.; Petronienė, S.; Gaižiūnienė, L.; Blanch, S.; Marbà-Tallada, A.; Brose, A. Towards active evidence-based learning in engineering education: A systematic literature review of PBL, PjBL, and CBL. Sustainability 2022, 14, 13955. [Google Scholar] [CrossRef]
  22. Wilkerson, B.; Trellevik, L.L. Sustainability-oriented innovation: Improving problem definition through combined design thinking and systems mapping approaches. Think. Ski. Creat. 2021, 42, 100932. [Google Scholar] [CrossRef]
  23. Teunissen, J.; Miller, G.; Casciani, D.; Colombi, C. Recalibrating fashion education in light of emerging fashion tech. In Proceedings of the 23rd Annual International Foundation of Fashion Technology Institutes (IFFTI), “Fashioning Resurgence—Our Time is Now”, Pearl Academy Creative Arts Education Society, New Delhi, India, 25 October 2021; pp. 10–16. [Google Scholar]
  24. Ferreira, V.G.; Canedo, E.D. Design Sprint in classroom: Exploring new active learning tools for project-based learning approach. J. Ambient. Intell. Hum. Comput. 2020, 11, 1191–1212. [Google Scholar] [CrossRef]
  25. Schulz, F.E.; Cunha, J. Women, Fashion Design and Ancestrality: Reflections on the Past and Future Possibilities. In International Fashion Design Congress; Springer: Cham, Switzerland, 2023; pp. 302–315. [Google Scholar] [CrossRef]
  26. Murni, A.; Sabandar, J.; Kusumah, Y.S.; Kartasamita, B.G. The enhancement of junior high school students’ abilities in mathematical problem solving using soft skill-based metacognitive learning. J. Math. Edu. 2013, 4, 194–203. [Google Scholar] [CrossRef]
  27. Cimatti, B. Definition, development, assessment of soft skills and their role for the quality of organizations and enterprises. Int. J. Qual Res. 2016, 10, 97. [Google Scholar]
  28. Schulz, B. The Importance of soft skills: Education beyond academic knowledge. Lang. Commun. 2008, 2, 146–154. [Google Scholar]
  29. Majid, S.; Liming, Z.; Tong, S.; Raihana, S. Importance of soft skills for education and career success. Int. J. Cross-Discip. Subj. Educ. IJCDSE 2012, 2, 1036–1042. [Google Scholar] [CrossRef]
  30. Snape, P. Enduring learning: Integrating C21st soft skills through technology education. Des. Technol. Educ. 2017, 22, n3. [Google Scholar]
  31. Poláková, M.; Suleimanová, J.H.; Madzík, P.; Copuš, L.; Molnárová, I.; Polednová, J. Soft skills and their importance in the labour market under the conditions of Industry 5.0. Heliyon 2023, 9, e18670. [Google Scholar] [CrossRef]
  32. Gilyazova, O.S.; Zamoshchansky, I.I.; Vaganova, O.I. Defining, classifying and developing soft skills in higher education: Competency-based and humanistic approaches. Univ. Soc. 2021, 13, 241–248. [Google Scholar]
  33. Ignacio, G.P.; Pilar, C.G.M.; Enrique, T.M. Impact of the sense of initiative and entrepreneurship competence on the entrepreneurial intention. Int. Entrep. Manag. J. 2023, 1–25. [Google Scholar] [CrossRef]
  34. Succi, C.; Canovi, M. Soft skills to enhance graduate employability: Comparing students and employers’ perceptions. Stud High. Educ. 2020, 45, 1834–1847. [Google Scholar] [CrossRef]
  35. Jones, D.E.; Greenberg, M.; Crowley, M. Early social-emotional functioning and public health: The relationship between kindergarten social competence and future wellness. Am. J. Public Health 2015, 105, 2283–2290. [Google Scholar] [CrossRef]
  36. Jeon, Y.; Jin, S.; Shih, P.C.; Han, K. FashionQ: An ai-driven creativity support tool for facilitating ideation in fashion design. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, Yokohama, Japan, 8–13 May 2021; pp. 1–18. [Google Scholar] [CrossRef]
  37. Shi, Y.; Gao, T.; Jiao, X.; Cao, N. Understanding design collaboration between designers and artificial intelligence: A systematic literature review. Proc. ACM Hum. -Comput. Interact. 2023, 7, 1–35. [Google Scholar] [CrossRef]
  38. Zhang, C.; Yao, C.; Wu, J.; Lin, W.; Liu, L.; Yan, G.; Ying, F. StoryDrawer: A child–AI collaborative drawing system to sup-port children’s creative visual storytelling. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems, New Orleans, LA, USA, 29 April–5 May 2022; pp. 1–15. [Google Scholar] [CrossRef]
  39. Huang, L.; Zheng, P. Human-computer collaborative visual design creation assisted by Artificial Intelligence. ACM Trans. Asian Low-Resour. Lang. Inf. Process. 2023, 22, 1–21. [Google Scholar] [CrossRef]
  40. Lee, G.; Kim, H.-Y. Human vs. AI: The battle for authenticity in fashion design and consumer response. J. Retail. Con. Serv. 2024, 77, 103690. [Google Scholar] [CrossRef]
  41. Liu, V.; Vermeulen, J.; Fitzmaurice, G.; Matejka, J. 3DALL-E: Integrating text-to-image AI in 3D design workflows. In Proceedings of the 2023 ACM Designing Interactive Systems Conference, Pittsburgh, PA, USA, 10–14 July 2023; pp. 1955–1977. [Google Scholar] [CrossRef]
  42. Haselberger, D.; Oberheumer, P.; Perez, E.; Cinque, M.; Capasso, D. Mediating Soft Skills at Higher Education Institutions. In Handbook of ModEs Project; Life Long Learning Programme: Brussels, Belgium, 2012. [Google Scholar]
  43. Kwon, E.; Rao, V.; Goucher-Lambert, K. Understanding inspiration: Insights into how designers discover inspirational stimuli using an AI-enabled platform. Des. Stud. 2023, 88, 101202. [Google Scholar] [CrossRef]
  44. Ma, J.J. Development of education for sustainable fashion design using a challenge-based learning approach. Int. J. Fash. Des. Technol. Educ. 2023, 16, 164–174. [Google Scholar] [CrossRef]
  45. Jung, D.; Suh, S.-E. Development of customized textile design using AI technology. J. Korean Soc. Cloth. Text. 2023, 47, 1137–1156. [Google Scholar] [CrossRef]
  46. Torres-Coronas, T.; Vidal-Blasco, M.A. Adapting a face-to-face competence framework for digital competence assessment. Int. J. Inf. Commun. Technol. Educ. 2011, 7, 60–69. [Google Scholar] [CrossRef]
  47. Conlon, J.; Gallery, C. Developing digital skills: A fashion business masterclass in virtual 3D prototyping with Style3D. Int. J. Fash. Des. Technol. Educ. 2023, 17, 76–85. [Google Scholar] [CrossRef]
  48. Tran, T.M. Integrating 21st Century Skills into Translation Classroom from Students’ Perspective. Int. J. TESOL Educ. 2023, 3, 64–78. [Google Scholar] [CrossRef]
  49. Choi, K.-H. 3D dynamic fashion design development using digital technology and its potential in online platforms. Fash. Text. 2022, 9, 9. [Google Scholar] [CrossRef]
  50. Tepe, J.; Koohnavard, S. Fashion and game design as hybrid practices: Approaches in education to creating fashion-related experiences in digital worlds. Int. J. Fash. Des. Technol. Educ. 2023, 16, 37–45. [Google Scholar] [CrossRef]
  51. Sun, L.; Zhao, L. Technology disruptions: Exploring the changing roles of designers, makers, and users in the fashion industry. Int. J. Fash. Des. Technol. Educ. 2018, 11, 362–374. [Google Scholar] [CrossRef]
  52. Kärnä-Behm, J. Learning generative design methods: Higher education students developing toolkits. Int. J. Art Des. Ed. 2022, 41, 577–588. [Google Scholar] [CrossRef]
  53. Hutson, J.; Robertson, B. Exploring the Educational potential of AI generative art in 3D design fundamentals: A case study on prompt engineering and creative workflows. Glob. J. Hum. -Soc. Sci. A Arts Humanit.-Psychol. 2023, 23, 485. [Google Scholar]
Figure 1. Schematic of research model for empirical, quantitative, and qualitative research methods.
Figure 1. Schematic of research model for empirical, quantitative, and qualitative research methods.
Sustainability 16 06973 g001
Figure 2. Schematic of the fashion textile pattern generation model (based on Jung and Suh [45]).
Figure 2. Schematic of the fashion textile pattern generation model (based on Jung and Suh [45]).
Sustainability 16 06973 g002
Figure 3. Student’s designing procedure and examples for fashion textile design from concept research, motif generation in Deep Dream Generator, and pattern generation in Wow Pattern to final virtual application in Zepeto.
Figure 3. Student’s designing procedure and examples for fashion textile design from concept research, motif generation in Deep Dream Generator, and pattern generation in Wow Pattern to final virtual application in Zepeto.
Sustainability 16 06973 g003
Figure 4. Student’s designing procedure and examples for fashion textile design from concept research, motif generation, and pattern generation to final virtual application in 3D CLO.
Figure 4. Student’s designing procedure and examples for fashion textile design from concept research, motif generation, and pattern generation to final virtual application in 3D CLO.
Sustainability 16 06973 g004
Figure 5. Distribution of students’ responses to AI use in the fashion textile design process.
Figure 5. Distribution of students’ responses to AI use in the fashion textile design process.
Sustainability 16 06973 g005
Table 1. Elements of soft skills, as clarified in the literature.
Table 1. Elements of soft skills, as clarified in the literature.
LiteratureElements of Soft Skills
Majid et al. [29]Teamwork and collaboration, decision-making, problem-solving, time management, critical thinking
Cimatti [27]Problem-solving skills, information analysis and synthesis, autonomous criticism, effective communication, life-long learning, teamwork, initiative, organization and planning
Snape [30]Communication (oral and written), critical thinking, problem-solving skills, work ethic, etiquette and good manners, organizational skills, courtesy, Inter- and intra-personal skills
Poláková [31]Problem-solving skills, communication skills, organizational skills, teamwork, leadership skills, creativity, analytical and critical thinking, value orientation, flexibility, initiative and engagement, learning skills, well-being focus, willingness to assume responsibility, emotional intelligence
Gilyazova et al. [32]Social and communicative skills (communicative skills, interpersonal skills, teamwork and leadership, social intelligence, responsibility, communication ethics)
Cognitive skills (critical thinking, problem-solving skills, innovative thinking, intellectual load management skills, learning skills, information skills, time management skills)
Personal attributes and emotional intelligence (emotional intelligence, integrity, optimism and positive thinking, flexibility, creativity, motivation, empathy)
Ignacio et al. [33]Positive attitude and initiative; communication and interaction; teamwork and collaboration; critical and analytical thinking or problem-solving, including risk assessment; creativity and innovation
Table 2. Allocation of teams for digital design experiences.
Table 2. Allocation of teams for digital design experiences.
Team NumberNumber of Team MembersTraditional Fashion Design Course ExperienceDigital Design Program AbilityDigital Fashion Textile Design ExperienceVirtual Fitting Program for Participants
Team 14Basic fashion design,
Fashion design inspiration,
Fashion Illustration,
Basic Draping,
Pattern Making
Adobe Illustration and PhotoshopNot experiencedZepeto
Team 23
Team 33
Team 43
Team 53Adobe Illustration and Photoshop,
3D CLO
ExperiencedZepeto,
3D CLO
Team 62
Team 73
Team 83
Team 93
Table 3. AI-based fashion textile design education course.
Table 3. AI-based fashion textile design education course.
CBL PhraseDS SessionSession GoalTask of the ClassTool
EngageStage 1: UnderstandOrientationUnderstanding the program and program goalSlide
Understanding AI design tools
(Deep dream generator, Wow pattern)
PC, laptop
Group organizationTeam setting, Pre-questionsPC, laptop
InvestigateStage 2: SketchConceptualizationResearch design concepts and motifs on open API sites on your ownPC, laptop
Think-aloud data
(Motif extraction)
Refine the image as per the design conceptDeep Dream Generator
Think-aloud data
(Pattern variation)
Coding the motif to a textile pattern and developing the textile design variationWow Pattern
Stage 3: DecideCommunication with team membersTextile pattern design process reviewPC, laptop
DecisionDecision on the final textile designPC, laptop
ActStage 4: PrototypeVirtual prototypesDecision on the final textile designPC, laptop
Apply the digital textile design to virtual clothingZepeto (mobile device), 3D CLO
Stage 5:
Validate
PresentationPresentation (fashion textile design and virtual prototypes)PC, laptop
Review and feedbackReview the process and final product with other teams
Post-education questions
PC, laptop
Note: AI, artificial intelligence; PC, personal computer; API, application programming interface.
Table 4. Soft skills developed at each stage of the process.
Table 4. Soft skills developed at each stage of the process.
Stages
1. Understand2. Sketch3. Decide4. Prototype5. Validate
Digital competence
Sense of initiative and entrepreneurship
Problem-solving
Visual communication
Table 5. Program satisfaction survey results.
Table 5. Program satisfaction survey results.
#Satisfaction SurveyMeanSV
#1Are you satisfied with the fashion textile design process and the AI curriculum?4.440.64
Essay Questionnaire
As a college student, it will be beneficial to learn about AI fashion textiles from the first year onward.
Prior knowledge of AI fashion textiles can provide students with an in-depth understanding of the development process.
Launching an item with a single outfit will be great.
Education in Photoshop (is necessary).
We should educate people about copyright and incorporate AI programs in courses that do not require significant levels of creativity.
Note: SV, standard deviation.
Table 6. Evaluation of the digital competence of soft skills.
Table 6. Evaluation of the digital competence of soft skills.
#QuestionMeanSV
#1Did the use of AI in fashion textile design complement your digital design skills?4.430.70
#2Did the use of AI in fashion textile design appropriately express your design intentions?4.210.89
#3Did the use of AI in fashion textile design improve the speed of your work?4.620.63
#4Did the use of AI in designing fashion textiles increase your work productivity by providing more design options?4.360.74
#5Did the use of AI in designing fashion textiles make your work easier?4.540.64
#6Did the use of AI in designing fashion textiles improve design quality?4.210.80
#7Can AI replace traditional digital design tools in fashion textile design?4.061.00
Note: SV, standard deviation.
Table 7. Open-ended questionnaire on digital competence.
Table 7. Open-ended questionnaire on digital competence.
Please describe in detail your satisfaction/dissatisfaction with AI use in textile design
SatisfiedI was pleased with how easy it was to create textiles and make the clothes I wanted.
I was satisfied that I could use AI to quickly and easily create textiles with the patterns and designs of my choice.
I liked that I could clarify how I wanted the image to look like in a short amount of time.
I liked the ease with which I could practice outside the class.
DissatisfiedAlthough I wanted an image of a pattern, it kept showing me an image of a person. I wrote “people” in the exclusion list; however, people kept coming up. Finally, the pattern worked out well; however, I think it came out slightly different from what I thought.
I would like to see more sizes to extract textiles from images.
Table 8. Evaluation of the sense of initiative and entrepreneurship.
Table 8. Evaluation of the sense of initiative and entrepreneurship.
#QuestionMeanSV
#1Do you think design methods using AI will inspire you to decide on a fashion textile design topic?4.290.72
#2Do you think designing with AI has helped you develop new fashion textile designs?4.370.74
#3Do you think it gives you an advantage in your design work compared to traditional fashion digital tools (e.g., Adobe Photoshop, and Textom Pro)?4.130.95
#4Do you think the fashion textile design method using AI will help you develop fashion products?4.290.78
#5Do you think the fashion textile design method using AI is economically viable for fashion product development?4.330.83
#6Do you think the fashion industry will require the skills of digital fashion textile design using AI in the future?4.290.87
#7Do you think AI-enhanced fashion textile design methods can replace traditional designer work?3.771.09
Note: SV, standard deviation.
Table 9. Open-ended questionnaire for sense of initiative and entrepreneurship.
Table 9. Open-ended questionnaire for sense of initiative and entrepreneurship.
Please provide a detailed analysis of the advantages and disadvantages of the design method using AI.
AdvantageDesign in less time.
Anyone can do it. However, you have to find the right image.
I think it is a great tool for fashion textile development because you can use it to create a pattern and apply it to create a design style quickly and easily to satisfy your needs.
Compared to other tools, it’s faster and, hence, it takes less time to come up with new designs.
Creating repeating patterns in Photoshop can be a time-consuming process; however, this time can be reduced.
DisadvantageThe lack of precision compared to Photoshop is slightly disappointing.
I think it may not produce the image pattern you want.
I can’t expect AI to be incredibly creative.
There’s a difference between what I think and what AI thinks.
It’s quick and easy to create a pattern; however, it can feel monotonous.
Table 10. Evaluation of communication.
Table 10. Evaluation of communication.
#QuestionMeanSV
#1Did you communicate with team members to plan the topic of fashion textile design using AI and understand the goals?4.350.80
#2Did AI-Based Fashion Textile Design Education course help you share your design process with your teammates?4.290.68
#3Do you think sharing feedback with other teams on your AI-driven fashion textile design development outputs helps your design work?4.360.74
Note: SV, standard deviation.
Table 11. Evaluation of problem-solving ability.
Table 11. Evaluation of problem-solving ability.
#QuestionMeanSV
#1Did this course help you in the development and application of fashion textile design?4.320.73
#2Do you think this course will improve your existing fashion textile design process?4.110.85
#3Did the implementation of the Zepeto process impact the original design intention of your textile work?3.880.75
#4Did the implementation of the 3D CLO process impact the original design intention of your textile work? (Group 5–9 only)4.360.63
Note: SV, standard deviation.
Table 12. Open-ended questionnaire for problem-solving ability.
Table 12. Open-ended questionnaire for problem-solving ability.
Please describe in detail the challenges you faced in developing fashion textiles using AI and the manner in which you overcame them.
Difficulties in selecting appropriate keywords and imagesI wanted the pattern image to be displayed; however, the image of a person kept appearing. Although I wrote “people” in the exclusion list, it kept displaying people. In the end, the pattern was generated; however, it didn’t look like what I thought it should look like.
In the Deep Dream Generator, I elaborated on the keywords and descriptions several times to get an accurate image suited to the concept.
During the development process, when I converted the image to an AI picture, I didn’t get the image I wanted as much as I thought. So, I kept running the AI and was able to get the image I wanted by using the correct name or command.
It was inconvenient that I couldn’t adjust the size of the repeating image at will; however, I was satisfied with the pattern I wanted after applying various types of patterns.
When I applied the textile to the design, I panicked because it wasn’t what I thought it would be. However, I could change the angle and image to get the design I wanted.
Difficulties of the processI think it will be easier to create textiles if there is a simpler path to create textiles. To overcome this challenge, I used relevant course materials and slowly followed the process to solve the problem.
Disappointment on outputDeveloping textiles using AI was a bit problematic in terms of creativity, and I was disappointed that I couldn’t come up with something that was significantly different from existing images.
If you expect this course to improve your current textile design process, please explain how it will do so.
Convenience and speedConvenience will be improved.
Convenience and speed will be improved.
There is a definite improvement in speed.
The development of AI textiles can benefit from improvements in the ability to efficiently create desired designs within existing frameworks to achieve intended purposes.
Variety of designsAI can develop various textile designs in different themes.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jung, D.; Suh, S. Enhancing Soft Skills through Generative AI in Sustainable Fashion Textile Design Education. Sustainability 2024, 16, 6973. https://doi.org/10.3390/su16166973

AMA Style

Jung D, Suh S. Enhancing Soft Skills through Generative AI in Sustainable Fashion Textile Design Education. Sustainability. 2024; 16(16):6973. https://doi.org/10.3390/su16166973

Chicago/Turabian Style

Jung, Dawool, and Sungeun Suh. 2024. "Enhancing Soft Skills through Generative AI in Sustainable Fashion Textile Design Education" Sustainability 16, no. 16: 6973. https://doi.org/10.3390/su16166973

APA Style

Jung, D., & Suh, S. (2024). Enhancing Soft Skills through Generative AI in Sustainable Fashion Textile Design Education. Sustainability, 16(16), 6973. https://doi.org/10.3390/su16166973

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop