On both digital capability and sustainability, advertising 4.0 service providers are pivotal for brands that compete. However, choosing the right partner requires a structured, data-driven approach. The purpose of this article is to propose a novel approach to assist sustainable fashion businesses in selecting the best advertising 4.0 services suppliers.
The study proposed a multi-step approach that included artificial intelligence (AI) -generated insights as well as expert validation. Generative AI developed the initial decision criteria, which served as a comprehensive and advanced starting point. These criteria were refined through expert focus groups to ensure their comprehensiveness and applicability. A multi-criteria decision-making technique, ordinal priority approach (OPA), utilized to rank the existing service providers. To validate the proposed approach and identify the best advertising 4.0 provider, we went through every step of the model and successfully ranked alternative advertising service suppliers in a fashion industry case study.
A user-friendly advertising 4.0 provider selection model was developed for the fashion industry with a sustainability competitive advantage, including Twenty-two comprehensive criteria for market knowledge, sustainability practices, and technical competence. Among the alternative service suppliers obtainable by the case company, this study revealed alternative 2 as the best-match advertising 4.0 partner. As a result of this case study applicability validation, the proposed framework is potentially beneficial for all industries, mainly the fashion sector, because it enables them to make informed decisions that are consistent with their strategic goals.
This article is among the pioneers to integrate artificial intelligence for criteria generation with expert validation and OPA optimization, producing a replicable generative AI-assisted MCDM framework. It promotes the concept of advertising 4.0 and contributes to the literature on service supplier selection, particularly in the sustainable fashion sector.
1. Introduction
Advancements in technology provide opportunities to utilize artificial intelligence (AI) to further improve decision-making practices (Choi et al., 2025; Marimira and Gumel, 2025), especially multi-criteria decision-making (MCDM) (Baydaş and Ersoy, 2025), with more effective solutions. Among existing AIs, generative AI is rapidly transforming decision-making across various domains (Albashrawi, 2025; Chuma et al., 2024). Thus, the objective of this research article is to propose a generative AI-powered MCDM framework for selecting advertising service providers. Companies in the current market should spend on both sustainability concerns and use of technological development (Sorooshian, 2025a); consequently, this article integrates technological capabilities and sustainability criteria to support marketing decision-making in the fashion industry.
This study follows a structured approach. Following this objective clarification, the article delves into a thorough literature review on Advertising 4.0, MCDM, and sustainable advertising to set the stage for its next proposal: an AI-powered approach to the selection of advertising 4.0 service providers. Through the use of a case study from the fashion sector, this article will examine the proposed approach, discussing its applicability and outcomes. Finally, the article proceeds with discussions and a conclusion.
2. Background
These days, the strong competition in the market has changed advertising into a highly important element of business strategy (Khan et al., 2024; Obi et al., 2023). In this regard, Khan et al. (2024) believed that advertising is highly effective in the growth and development of businesses. Due to the transformative nature of the marketing environment, it is necessary for companies to employ creative strategies to attract the attention of customers in such a way that they would remain loyal to the intended products (Lee and Cho, 2020). In addition to increasing the rate of product sales, advertising livens up an interactive sense with products and brands among customers. The effectiveness and efficiency of advertisement assume paramount importance in the improvement of any business since it can create a good market for the business products (Etale, 2015). It also directs the way customers and consumers may perceive the products and brands, which finally leads to business promotions and lasting viability.
In this light, Industrial Revolution 4.0 (Industry 4.0) and its technologies have made significant changes to the advertisement approaches and methods (Lee and Cho, 2020; Rosário and Dias, 2022). Today, the advancement of technology has had a significant impact on businesses (Abirami et al., 2023), increasing reliance on it for real-world problem-solving. As time passes, technology and communication evolve, influencing how communicators work, messages are sent, recipients receive messages, feedback is received and media functions (Geiβ et al., 2013). Societies expect immediate, practical solutions, requiring technology to be more innovative and meet user needs (Rahmat, 2021). Technology and the Internet have accelerated the evolution of marketing tools (Jackson and Ahuja, 2016; Dsouza and Panakaje, 2023), including advertising. In fact, old advertising strategies have undergone big transformations by integrating Industry 4.0 technologies, including big data analytics, cloud computing, AI and extended reality. As a result, advertising is now more effective and efficient than before (Vittala et al., 2024; Geng, 2022; Turhan, 2022). Industry 4.0 technologies benefit the marketing industry (Mukhopadhyay et al., 2024) and with utilization of these technologies, advertising 4.0 introduced.
Advertising 4.0 is delineated in this article for the first time, though the concept has been used by other researchers (e.g. (Ramachandran et al., 2025; Sajan and Giri, 2025; Theodorakopoulos et al., 2025)). The term signifies the transformative evolution in advertising strategies, propelled by the incorporation of Industry 4.0 technologies. While Advertising 4.0 is closely linked with digital marketing, it represents a specialized approach within digital advertising. It leverages advanced Industry 4.0 technologies to deliver highly personalized, interactive, and context-aware consumer experiences (Rosário and Dias, 2022). Instances for advertising 4.0 are, for one, companies, with the goal of enabling targeted engagement at an individual level, use AI and big data analytics to deliver hyper-personalized messages (Singh and Kakkar, 2025; Singh and Kaunert, 2024; Santos, 2021),¨ or real-time advertising strategy adjustment (Balamurugan, 2024), or offering dynamic pricing systems, price changes that is based on individual's interactions or behaviors (Kopalle et al., 2023), to display personalized advertisements (Immadisetty, 2025; Awais, 2024; Pradhan et al., 2025). Marketing communications or advertisements that enable consumers to use extended reality technologies for interacting with products in simulated environments (Scholz and Duffy, 2018; Ozturkcan, 2021) is another example of it. Further instances include, advertising in the metaverse (Eyada, 2023; Dwivedi et al., 2023; Levin and Osborne, 2025), utilizing IoT (Internet of Things)-connected beam projectors to deliver customized advertising content based on business hours and location data (Kwon, 2024), and deploying text-, voice- or video-based chatbots for personalized interactions within messaging applications or social media (Torous et al., 2021; Zumstein and Hundertmark, 2017).
Although businesses face numerous challenges in maximizing this modern advertising outcomes, it has become indispensable in today's marketing strategies (Prihatiningsih et al., 2025). Haleem et al. (2022), for instance, argue that the use of big data analytics and AI technology in advertising improves campaign efficiency. Besides, the algorithms arranged according to AI have the capability of predicting consumer intents and tastes and promoting advertisement status (Haleem et al., 2022). In this regard, Wu et al. (2022) and Sorooshian (2024) argue that extended realities and metaverse have the potential to bring such brand experiences into reality that would engage customers and consumers. Jayawardena et al. (2023), for another instance, illustrate the persuasion effects of advertisements with extended reality technologies. Nonetheless, modernization has all come into play as a result of reaction to customers' taste and response and market demand for precision and interaction. Industry 4.0-related technologies and power in the analysis of customer behavior are among the most characteristic qualities of advertising 4.0. As a result, the product has been the sound integration of modern digital platforms (Juska, 2021; Kumar and Gupta, 2016).
These technological transformations have paved the way for enterprises and companies to achieve their desirable goals and objectives more effectively and efficiently. Moreover, real-time data analysis provides grounds for the swift adjustment of companies and, accordingly, more return on investment (Snyder and Garcia-Garcia, 2016; Yuan et al., 2013). When it comes to retail, Malekpour et al. (2023) predict that digital channels will have an even bigger impact on emerging markets. It is noteworthy that such advertisements are put into practice in different industries, such as the fashion industry, in order to create interactive product experiences. According to Shah and Nasnodkar (2023), and Madhura and Panakaje (2022), the modern era has witnessed special benefits to the fashion industry as a result of advanced advertising. It is required for fashion products to keep updated regarding the latest market developments. For this reason, they may take advantage of social media data analytics, AI-powered trend analysis, and influencer marketing to have an active, interactive connection with customers and bring a competitive advantage (Rathore, 2019). The powerful visual attractiveness of fashion items is desirable for digital advertising platforms since they can more efficiently involve customers with the intended content and new collections. Fashion brands make use of modern advertising approaches to influence people throughout the world and create powerful brand identities. In this way, they can redefine and modify their policies and products in response to customers' wants and market demand.
Thereupon, it's possible that consumers' growing awareness of ethical and environmental concerns is driving the rise of the sustainable fashion industry. These days, customers and consumers value the companies that feel more committed to the environment (Haddock-Fraser and Tourelle, 2010); and contemporary advertising contributes significantly to the advancement of sustainable fashion by informing consumers about the benefits of sustainable practices and highlighting the efforts made by brands to reduce their environmental footprint (Mohr et al., 2022; Evans and Peirson-Smith, 2018). The importance of sustainable fashion businesses has grown significantly as consumers gravitate toward brands that align with their values of ethical manufacturing and sustainability. This transition gives fashion brands a competitive advantage if they prioritize sustainability (Öndoğan et al., 2022), especially in developing countries where sustainability issues are still relatively new but rapidly growing (Nguyen et al., 2021). Nowadays, an increasing population of companies and customers in developing countries is getting informed of the long-term advantages of making use of sustainable criteria, which, in itself, places a greater emphasis on sustainability. The implementation of sustainability in the fashion industry considers environmental criteria and also appeals to customers. This leads to the improvement of loyalty and competitiveness in the intended brand. Such businesses are distinctively different from other competing businesses since they enjoy the competitive advantage of sustainability (Öndoğan et al., 2022; Nguyen et al., 2021).
It seems highly essential for businesses and companies to come with the best possible decisions and policies in sending their desired message to the customers by means of advertising services (Komulainen et al., 2007). Such a decision is so complex that multiple criteria should be taken into account. Looking at the related literature, one notices the existing gap between the current advertising strategies and choosing the eligible providers of advertising services.
The selection of an advertising provider is a systematic process that prioritizes the identification of a cooperative supplier capable of delivering quality service (Sorooshian, 2025b). In the selection of Advertising 4.0 service providers, companies should necessarily follow an inclusive approach in which they can influence the target customers through the primary competition and obtain good outcomes. In terms of selecting Advertising 4.0 service providers, this study aims to develop a comprehensive framework for decision-making by multiple criteria and put it to the test in the scope of the fashion industry in developing countries. MCDM appeared as a promising approach for supplier selection in many industries (Sahoo et al., 2024; Khulud et al., 2023). Nonetheless, their application capabilities in marketing and advertising service provider selection have been significantly overlooked, particularly in the fashion industry: On 14th April 2025 the author performed a literature search in the Scopus database, the largest scientific database (Sorooshian and Ebrahim, 2023) and one of the main existing ones (Sorooshian et al., 2023), using a search string that combined keywords related to multiple criteria decision-making, specifically “multiple criteria decision-making,” “multi-criteria decision-making” and “MCDM,” with terms in the marketing and advertising domain, along with a focus on “fashion.” The query targeted documents where these terms appeared in the title, abstract, or keywords and was restricted to both Business and Decision Sciences subject areas. However, the search did not yield any relevant studies. In this study, to fill the literature gap, we use an innovative AI-powered approach to help with the selection of advertising 4.0 providers for the sustainable fashion industry's MCDM process. As a result, the contributions of this article are:
This study is one of the first to address the problem of advertising services supplier selection. Particularly, the authors present a first-of-its-kind MCDM framework for the sustainable fashion industry to select advertising 4.0 service providers. To validate our approach, we used a case study on a sustainable fashion company in a developing country.
A recent MCDM literature analysis (Kumar and Pamucar, 2025) suggested that new studies need to focus on finding ways for integrating MCDM with cutting-edge technologies consisting of AI. Addressing this need made our proposed approach become one of the pioneers. This research uses the data-driven power of AI to improve classic MCDM procedures. As, in many cases, businesses must make decisions in fields where they lack expertise, this study, particularly in cases where the level of competence for making the decision is insufficient, proposes a solution: an AI-based data-driven MCDM approach.
Based on this background, the following section will explain our methodology to create a decision-making system that is specifically designed to handle the issues that have been identified.
3. Methodology
As a consultant, one of the authors of this article worked with a sustainable fashion business in a developing country, a seller of sustainable clothing and accessories in a developing country, that wanted to improve their advertising campaigns by utilizing modern technologies. For this company, whose main strength and market promise lie in its eco-credentials, choosing a modern advertising service provider was a concern, as it lacked sufficient experience with technology-based advertising systems. They had marketing and advertising experience and expertise, but were experiencing their first advertising. 4.0 attempt. To balance their lack of experience and knowledge, after reviewing the literature background in Section 2, we offered the company a data-driven decision-making approach for choosing an advertising service supplier.
Were the participants who provided consent for the study. Figure 1 visualizes the proposed seven-step approach that we offered to the fashion case business.
The figure presents a horizontally oriented, stair-step style flowchart composed of seven rectangular boxes aligned diagonally from the top left corner toward the bottom right corner. Each box is filled with different shades that become progressively darker as the flow advances. From each box, a downward arrow arises and points downward to the next box. The sequence begins with the box containing the label “Step 1. Forming an expert panel”. An arrow arises from this box and points downward to the next box labeled “Step 2. Generates preliminary criteria by A I”. Moving forward, the boxes are labeled “Step 3. Validating the criteria by the panel”, “Step 4. Pairwise weighting”, “Step 5. Computing the criteria weights”, “Step 6. Scoring supplier alternatives”, and “Step 7. Selecting the best supplier”.Multiple criteria selection of advertising 4.0 suppliers
The figure presents a horizontally oriented, stair-step style flowchart composed of seven rectangular boxes aligned diagonally from the top left corner toward the bottom right corner. Each box is filled with different shades that become progressively darker as the flow advances. From each box, a downward arrow arises and points downward to the next box. The sequence begins with the box containing the label “Step 1. Forming an expert panel”. An arrow arises from this box and points downward to the next box labeled “Step 2. Generates preliminary criteria by A I”. Moving forward, the boxes are labeled “Step 3. Validating the criteria by the panel”, “Step 4. Pairwise weighting”, “Step 5. Computing the criteria weights”, “Step 6. Scoring supplier alternatives”, and “Step 7. Selecting the best supplier”.Multiple criteria selection of advertising 4.0 suppliers
Step 1: It is essential for business groups to have a decision-making panel when making decisions in groups. However, panels may choose to include organization's external experts or rely solely on internal sources when making their nominations. We, however, invited the case company to nominate individuals who can act as their marketing expert. Due to the exploratory nature of the study, resource constraints, and business strategic sensitivity, the decision-making panel consisted exclusively of four senior staff from the company's management and marketing team who provided consent to participate in this study. This panel of experts was agreed upon because it also leverages internal strategic alignments for the advertising supplier selection.
Step 2: We brought the nominated four panel members to help us maximize the potential of generative AI. While there are options for AI tools, we decided to use ChatGPT (https://chatgpt.com/), a large language model-based generative AI model created by OpenAI (https://openai.com/), to generate a comprehensive set of decision criteria. This AI chatbot is revolutionizing now (George and George, 2023; Shi et al., 2024). George (George and George, 2023) explains that businesses, by utilizing ChatGPT, as an efficient natural conversational chatbot, can reduce their costs, improve customer service and enhance their understandings of customer behavior. All the panel members had experience using ChatGPT before this practice, so no training was required.
ChatGPT is an advanced chatbot built with a generative pre-trained transformer (GPT) architecture. The GPT is an effective tool for benchmark generation and ideation because it, as Yekta explains (Yekta, 2024), can carry out complex tasks by integrating data from multiple sources. ChatGPT's responsibilities, thus, include providing relevant information, recommending probabilistic metrics, and generating insights from massive amounts of data, as explained by Ray (2023). These feature is intended to help people make data-driven decisions (Choudhury and Shamszare, 2023). ChatGPT, hence, contributed to the development of the first set of criteria for evaluating advertising 4.0 service providers as part of our research.
The four panel members were well-versed in general marketing and management, but they were not fully confident in the concept of advertising 4.0. As a result, the ChatGPT generative AI utilization was chosen for use in sustainable fashion advertising because of its ability to synthesize existing information and generate new, interconnected factors that may influence decisions. In this step, thus, each of the company's panel members used ChatGPT to identify decision criteria that are critical for advertising 4.0 providers. The formulation of a prompt to communicate with the chatbot can occasionally dictate its effectiveness. In order to address this issue, since the team had enough experience using ChatGPT, we enlisted the expertise of our panel members to generate and refine their own set of ChatGPT inquiries. By employing this approach, we successfully encompassed a diverse range of ChatGPT's functionalities and outcomes, all the while making progress toward our shared goal. As a result, any questions that panel members felt were important for improving the list of criteria, and/or their understanding of any criterion, could be directed to ChatGPT. The final criteria had to be thorough and applicable, so this was completed. There were several good reasons to use ChatGPT on its own, utilized by each and every panel member individually, to create benchmarks.
ChatGPT, a sophisticated generative AI model, can efficiently handle and analyze large amounts of data. Its strength lies in the unique set of decision criteria it generates by combining data from various fields, which may be difficult for humans to identify. ChatGPT was used to generate an all-inclusive set of Advertising 4.0 metrics, including insights from multiple sources. Our decision-making model was made more thorough and relevant by incorporating these initial AI-generated lists, which included a wide range of relevant data-driven factors and ensured that the criteria of interest were up to date.
Step 3: The panel members conducted a focus group review of the proposed ChatGPT criteria. The authors of this article led the panel with the common goal of developing a set of criteria that would meet the unique needs of the sustainable fashion sector while also being comprehensive and free of duplication. After using the focus group method to combine their individual lists, the panel arrived at a final set of criteria. To provide the best possible definition and terminology used for the criteria list, it was also permitted to refer to ChatGPT for help.
The review process applied by this group provided the possibility of cross-validating AI-generated criteria, which finally led to an inclusive, error-free list of criteria. In designing this list, the mentioned group provided the grounds for the precise discussion of all the criteria one by one, which minimized the potential threats, such as confusion and duplication. The appropriate theoretical and practical features were applied to the selection of criteria, where the difficulty of this collective task, as a critical challenge, was evident. This panel acted as a forum for achieving a final agreement and this led to the assignment of credit to member opinions in coming to a final list of criteria.
Step 4: The next step was the ranking of each criterion by the panel members to gain the likes and dislikes of the individuals directly involved in the decision-making process. These criteria were matched with the conditions of a case study company. The opinions and experiences of each of the panel members drastically influenced the final selection of the list of criteria. Each of the panel members was separately asked to rank the criteria so that different opinions and perspectives could be manifested in the final framework. In this way, varying priorities of the panel members were identified, and we came to a better understanding of the importance of each criterion. An individualized approach in rating was practiced, integrating the panel's opinions into MCDM. Accordingly, the criteria were ranked based on theoretical and practical relevance to the case company conditions.
Step 5: This stage is to calculate the weights of the criteria using MCDM methods. It is, however, noteworthy that both traditional and modern methods are present in MCDM at the same time, whereas the exact number of techniques has remained unidentified because of the continuous emergence of integrative approaches and progressive transformation of techniques (Mahmoudi et al., 2021). Various MCDM techniques result in different outcomes owing to the differences in computational intensity, assumptions, and limitations (Wu and Abdul-Nour, 2020; Sorooshian et al., 2023). In this regard, it is essential that the possible ambiguities in modern decision-making be eliminated and, therefore, old methods should be left out and modern ones should be put into practice instead. The new ones enjoy some benefits that allow them to manage the complexity of the problems and issues and come to educated decisions owing to the progress made in MCDM. Therefore, the ordinal priority approach (OPA) was used in this research, which is a modern method of multi-criteria decision making (Ataei et al., 2020).
In this light, Pan et al. (2024) have reported that it is possible to obtain more reliable responses by OPA in comparison with other MCDMs. Mahmoudi and Javed (2022) believe that the advantages of OPA have converted it into a versatile approach whose application can solve supplier (including service supplier) selection problems. Sorooshian (2025b) verified OPA's feasibility for marketing 4.0 service supplier selection. Indeed, OPA is a relatively new decision-making method that can tackle the MCDM-related problems (Pan et al., 2024). Moreover, it, due to its unique strength in systematically incorporating ordinal data from expert judgments, thereby significantly reducing subjective bias (Mahmoudi and Javed, 2023), is deliberately chosen. The following presents the interpretation of OPA constituent procedures following the explanation made by Hashemkhani Zolfani et al. (Hashemkhani Zolfani et al., 2022). Table 1 displays these procedures, which include the necessary sets, indexes, variables and decision-making criteria. As previously stated, OPA was used to build and solve the linear programming model, where decision panel members used cumulative weights to rank the criteria. As a result, we were able to objectively identify which criteria were most important when selecting advertising 4.0 providers. In OPA, the decision-making panel should identify which selection criterion they prefer the most. A linear programming model (equation 1), then, should be built using the data gathered from the panel.
Sets, indexes and decision variables
| Sets | |
|---|---|
| I | Set of n panel members |
| J | Set of m criteria |
| K | Set of alternatives |
| Indexes | |
|---|---|
| i ∀i ∈ I | Index of panel members |
| j ∀j ∈ J | Index of criteria |
| k ∀k ∈ K | Index of preference of the attributes |
| p | Order of criteria j |
| Decision variables | |
|---|---|
| Z | Objective function |
| Wace(p) | Weight (importance) of jth criterion to ith panel members at pth rank |
| Aijk | The kth alternative |
Max Z.
Subject to:
When the built-model of equation 1 is solved, the weight of the criteria can be calculated using equation 2.
Step 6: Similar to the stage 4 process, the panel members have provided scopes for existing advertising service supplier alternatives. There were three alternative advertising providers for the case fashion company's decision-making, so the panel members were asked to provide scores for each criterion, considering each and every alternative advertising provider.
Step 7: The final stage is the calculation of the final scores of alternatives, according to equation 4, and ranking them.
Results of our case study will validate the applicability and feasibility of the proposed approach (Ramezanzade et al., 2021), thereby confirming the attainment of our research objective.
4. Results
Following the nomination of the decision panel in step 1 and step 2 for AI-powered criteria identification, a focused group was formed in step 3. The focus group's results, Table 2, showed that when selecting advertising 4.0 providers for the sustainable fashion business case, it is essential to evaluate them against a set of comprehensive 22 criteria ensuring they meet both technological and sustainability goals. Here are the decision criteria that can guide the selection process:
Selection criteria
| Criterion | Index |
|---|---|
| Technological Capability: Proficiency in AI, big data analysis, and extended reality, along with the robustness and reliability of technological infrastructure | C1 |
| Experience in Sustainable Fashion: Demonstrated expertise and track record in the sustainable fashion industry, including understanding of target markets and consumer behaviors in developing countries | C2 |
| Cost Efficiency: Cost-effectiveness of services, ability to deliver high ROI and provision of high-quality services at a reasonable cost | C3 |
| Sustainability Practices: Commitment to minimizing environmental impact, ethical practices, transparency and sustainable practices within the provider's operations | C4 |
| Adaptability and Customization: Ability to tailor advertising strategies to specific brand needs and adapt to changing market conditions and emerging trends | C5 |
| Market and Local Knowledge: Depth of knowledge about the fashion industry, current market trends and understanding of economic, social factors and cultural contexts | C6 |
| Innovation and Creativity: Track record in innovative advertising campaigns, quality and originality of content and creativity in visual design | C7 |
| Data Analytics Expertise: Ability to analyze consumer behavior and campaign performance in real-time and utilize data for campaign adjustments. Robust mechanisms for measuring campaign performance and providing detailed, actionable reports | C8 |
| Digital Platforms Integration and Multichannel Execution: Seamless integration of digital platforms and proficiency in executing campaigns across multiple channels | C9 |
| Experience and Reputation: Established experience and credibility within the industry and among clients | C10 |
| Communication Skills: Effectiveness in communicating strategies, results and feedback with clients | C11 |
| Client Testimonials and Case Studies: Feedback and case studies from previous clients demonstrating success | C12 |
| Brand Alignment: Alignment of the provider's values and practices with the sustainable and ethical values of the fashion brand | C13 |
| Project Management Skills: Ability to manage projects efficiently, meet deadlines and handle resources effectively | C14 |
| Audience Engagement and User Experience: Strategies for engaging with the target audience and expertise in creating user-friendly digital experiences | C15 |
| Legal and Regulatory Compliance: Encompasses adherence to local and international advertising regulations and standards | C16 |
| Cross-Cultural Competence: Ability to navigate different cultural contexts, especially in developing countries | C17 |
| Security and Privacy: Ensuring the security and privacy of consumer data and compliance with data protection regulations | C18 |
| Long-Term Partnership Potential and Scalability: Potential for establishing long-term partnerships and ability to scale services according to brand growth | C19 |
| Visual and Aesthetic Quality: High standards in the visual and aesthetic quality of advertising materials | C20 |
| Influencer and Partner Network: Access to a relevant network of influencers and industry partners | C21 |
| Crisis Management and Contingency Planning: Ability to handle unforeseen issues during the campaign and approach to contingency planning | C22 |
The decision panel members are also presumed to possess comparable weight in the decision-making process, as in the 4th step, and their data is treated with equal regard. The 1–5 scale was utilized, with 5 representing the highest weight and 1 representing the lowest weight. Table 3 presents the color-scaled positions of the panel members regarding how to rank each of the criteria.
Opinions on criteria ranking
| Criteria | Member 1 | Member 2 | Member 3 | Member 4 |
|---|---|---|---|---|
| C1 | 3 | 4 | 3 | 4 |
| C2 | 4 | 3 | 3 | 3 |
| C3 | 5 | 4 | 5 | 5 |
| C4 | 5 | 5 | 5 | 5 |
| C5 | 4 | 4 | 4 | 3 |
| C6 | 4 | 5 | 4 | 4 |
| C7 | 4 | 4 | 5 | 4 |
| C8 | 3 | 4 | 3 | 3 |
| C9 | 4 | 3 | 5 | 3 |
| C10 | 4 | 4 | 4 | 3 |
| C11 | 4 | 4 | 5 | 5 |
| C12 | 3 | 2 | 3 | 3 |
| C13 | 4 | 4 | 4 | 3 |
| C14 | 4 | 4 | 5 | 4 |
| C15 | 2 | 3 | 3 | 2 |
| C16 | 3 | 2 | 3 | 2 |
| C17 | 3 | 3 | 2 | 2 |
| C18 | 2 | 2 | 3 | 3 |
| C19 | 3 | 4 | 5 | 4 |
| C20 | 4 | 4 | 4 | 4 |
| C21 | 4 | 4 | 3 | 3 |
| C22 | 1 | 2 | 2 | 2 |
In step 5, subsequently, the criteria weights were overviewed through the utilization of the OPA linear optimization model as outlined in appendix, equation 1A. The following equations 2 −23A of the appendix must be taken into account in order to derive the ultimate weights of the criterion after the aforementioned model has been solved.
A JavaScript web-based software (https://ordinalpriorityapproach.com/index.php?s=4-opa-software) is developed and recommended for OPA calculation in MCDM (Mahmoudi et al., 2023). This OPA Solver provides the results as follows: W (C1): 0.029128; W (C2): 0.037277; W (C3): 0.028712; W (C4): 0.019974; W (C5): 0.035543; W (C6): 0.036098; W (C7): 0.023719; W (C8): 0.047160; W (C9): 0.033497; W (C10): 0.040571; W (C11): 0.028712; W (C12): 0.080103; W (C13): 0.060857; W (C14): 0.032457; W (C15): 0.058256; W (C16): 0.100909; W (C17): 0.052015; W (C18): 0.050281; W (C19): 0.032041; W (C20): 0.035890; W (C21): 0.031729; W (C22): 0.105070.
Figure 2 transformed this information into a visual representation.
The horizontal axis displays criterion labels of each bar labeled from left to right as follows: “C 22”, “C 16”, “C 12”, “C 13”, “C 15”, “C 17”, “C 18”, “C 8”, “C 10”, “C 2”, “C 6”, “C 20”, “C 5”, “C 9”, “C 14”, “C 19”, “C 21”, “C 1”, “C 3”, “C 11”, “C 7”, and “C 4”. The vertical axis is unlabeled in the figure but represents the numerical weight values shown above each bar. The bars decrease in height from left to right. The data for the bars is as follows: C 22: 0.10507. C 16: 0.100909. C 12: 0.080103. C 13: 0.060857. C 15: 0.058256. C 17: 0.052015. C 18: 0.050281. C 8: 0.04716. C 10: 0.040571. C 2: 0.037277. C 6: 0.036098. C 20: 0.03589. C 5: 0.035543. C 9: 0.033497. C 14: 0.032457. C 19: 0.032041. C 21: 0.031729. C 1: 0.029128. C 3: 0.028712. C 11: 0.028712. C 7: 0.023719. C 4: 0.019974.Order of criteria weights
The horizontal axis displays criterion labels of each bar labeled from left to right as follows: “C 22”, “C 16”, “C 12”, “C 13”, “C 15”, “C 17”, “C 18”, “C 8”, “C 10”, “C 2”, “C 6”, “C 20”, “C 5”, “C 9”, “C 14”, “C 19”, “C 21”, “C 1”, “C 3”, “C 11”, “C 7”, and “C 4”. The vertical axis is unlabeled in the figure but represents the numerical weight values shown above each bar. The bars decrease in height from left to right. The data for the bars is as follows: C 22: 0.10507. C 16: 0.100909. C 12: 0.080103. C 13: 0.060857. C 15: 0.058256. C 17: 0.052015. C 18: 0.050281. C 8: 0.04716. C 10: 0.040571. C 2: 0.037277. C 6: 0.036098. C 20: 0.03589. C 5: 0.035543. C 9: 0.033497. C 14: 0.032457. C 19: 0.032041. C 21: 0.031729. C 1: 0.029128. C 3: 0.028712. C 11: 0.028712. C 7: 0.023719. C 4: 0.019974.Order of criteria weights
The criteria can be organized in this way, allowing the advertising service provider selection process to focus on the most important aspects. While calculating the OPA results, the OPA solver additionally generates Kendall's W (Ci), also known as the rank coefficient, which quantifies how congruent two rankings are. Values near 1 indicate strong agreement, while values near −1 indicate strong disagreement (Kendall, 1938). As shown in Table 4, all criteria resulting from the focus group are free of disagreement.
Confidence level measures
| Criteria | C1 | C2 | C3 | C4 | C5 | C6 | C7 | C8 | C9 | C10 | C11 | C12 | C13 | C14 | C15 | C16 | C17 | C18 | C19 | C20 | C21 | C22 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Kendall's W (Ci) | 1 | 1 | 0.75 | 1 | 0.942 | 0.365 | 1 | 0.7 | 0.5 | 1 | 0.25 | 0.813 | 0.813 | 0.75 | 0.75 | 0.25 | 0.25 | 0.438 | 0.25 | 0.438 | 0.25 | 0.712 |
Next, because three advertising 4.0 providers were available for ranking, the decision panel was asked to provide scores for each alternative for each criterion in step 6. The ultimate stage result, as illustrated in Figure 3, involved computing the alternatives' final scores and assigning a ranking to each in accordance with the scores. As can be seen in the figure, one of the advertising service suppliers, alternative 2, had the highest score of 0.385, making it the preferred option as the advertising provider for the case company. Alternative 1 appears to be the least optimal choice when compared to other alternatives. This was a thorough and methodical approach to advertising 4.0 provider selection, according to the case company representatives as well.
The horizontal axis of the graph ranges from 0 to 0.4 in increments of 0.1 units. The markings on the vertical axis from top to bottom are “Alternative 3”, “Alternative 2”, and “Alternative 1”. The data from the bars is as follows: Alternative 3: 0.342534. Alternative 2: 0.385602. Alternative 1: 0.271864.Comparison of Alternatives scores
The horizontal axis of the graph ranges from 0 to 0.4 in increments of 0.1 units. The markings on the vertical axis from top to bottom are “Alternative 3”, “Alternative 2”, and “Alternative 1”. The data from the bars is as follows: Alternative 3: 0.342534. Alternative 2: 0.385602. Alternative 1: 0.271864.Comparison of Alternatives scores
5. Discussion
This study proposed a seven-step framework for multiple-criteria selection of advertising 4.0 service suppliers. This framework, which uses generative AI to generate initial criteria and then refines them with decision panels, contributes to the existing literature. Recent literature reviews (Sedkaoui and Benaichouba, 2024; Salih et al., 2025) have emphasized the transformative potential of generative AI in augmenting human creativity and decision-making. They also show that there is a limited understanding of how it can systematically improve decision frameworks. Alam and Khan (2024) have also reported a lack of a decision framework that aims to bridge the gap between AI systems and human decision-makers. In response, this study presents decision-making with AI-human collaboration, a pioneering theoretical synthesis by integrating generative AI with MCDM theory. The utilization of generative AI aligns with the ongoing industrial revolution (Industry 5.0) objectives of sustainability and human-centric innovation, as highlighted by Salih et al. (2025), so this study is a step forward for the Industry 5.0 movement. Moreover, by filling a gap and addressing the need for modern advertising services for businesses, this study significantly contributed to the marketing literature. This is because previous studies have identified the advertising industry as benefiting significantly from technological advancement (Vittala et al., 2024; Geng, 2022; Turhan, 2022; Jayawardena et al., 2023; Haleem et al., 2022), but there is a lack of research on the systematic selection of service suppliers (Sorooshian, 2025b).
As an empirical insight, our case study confirmed the framework's feasibility in the fashion industry. Besides, the finding demonstrate interdependency of technological capabilities and sustainability criteria, advancing traditional supplier selection models seeing these aspects separately (e.g. (Falatoonitoosi et al., 2013; Yadav and Sharma, 2016)). These results indicate that sustainable fashion brands may benefit from clearly emphasizing their commitment to sustainability and digital innovation in their advertising strategies. These strategies have the potential to positively impact consumer attitudes, resulting in better brand differentiation. Sustainability commitments strongly influence competitive advantage in fashion industries (Resta et al., 2018; Razzak, 2023); technological capability is also identified as a connected aspect, supported by research indicating that technology and digital innovation drive advancements in sustainability capability (Huynh, 2022; Ikram, 2022). In this study, the traditionally pivotal criterion, cost efficiency decision criterion ranked significantly lower than some other factors, such as legal and regulatory compliance, in direct consideration. This confirms (De Ponte et al. (2023) saying that commitment to policies and regulations provides a competitive advantage to fashion businesses both in terms of brand image and cost reduction through increased production efficiency. This suggests that sustainable fashion businesses may prioritize long-term strategic alignment over immediate cost-saving.
By means of ChatGPT, the primary decision criteria were obtained and these initial criteria constituted the foundation for the provision of a final criteria list by a focus group. Twenty-two criteria were finally developed that encompassed a variety of subjects, such as sustainability practices, knowledge, and technical competency. Crisis management and contingency planning, as well as legal and regulatory compliance, are highly weighted by the case company panel. The literature confirms these as marketing failure risk management is critical to success of marketing campaigns (Sorooshian, 2025b; Tauro et al., 2021); besides, compliance with legal and regulatory compliance is a distinguishing feature of fashion advertising ecosystems (Saha, 2023; Cerchia and Piccolo, 2019) and should be among the primary selection criteria when hiring cutting-edge service providers for cross-border or culturally sensitive campaigns. In the case study, alternative advertising supplier 2 outperformed the others. A deeper examination reveals that Alternative 2 ranked significantly higher on critical criteria such as market and local knowledge, and experience in sustainable fashion. This emphasizes the importance of long-term development and professional continuity over time (Sun et al., 2025).
For industry practitioners, the framework design enables businesses to integrate their technological and environmental objectives into a single cohesive system, laying the groundwork for future sustainable advertising strategy innovations. Consequently, the procedure helped with the systematic optimal choice of options. Managers are assisted in obtaining their strategic and technological goals using these research findings, where they can appropriately choose advertising 4.0 services suppliers within the comprehensive framework obtained. Due to the dynamic nature of the advertising industry and sustainability, the active companies in developing countries are likely to take the most advantage of this framework. One of the abilities of generative AI, including ChatGPT, is to convert large amounts of data into useful information (Hassani and Silva, 2023). The proposed seven-step framework, correspondingly, allows businesses to easily update their criteria based on generative AI data-driven support, as well as their marketing strategies, to meet changing consumer needs and market trends. Accordingly, companies and businesses can make educated decisions to approach their important goals as the main emphasis of the framework is placed upon the important criteria toward the achievement of technological and sustainability goals.
6. Conclusion
The study addressed the shortcomings of advertising service supplier selection. Based on the study's findings, a comprehensive framework was created to assist sustainable fashion businesses in developing countries in selecting the most appropriate advertising 4.0 services supplier. It aimed at making considerable theoretical and practical contributions to the fashion industry. A practical application for businesses committed to sustainable fashion was offered in this study by highlighting the criteria required to meet technical and sustainability goals. This research's decision-making approach can be used not only by businesses involved in the sustainable fashion industry to choose advertising 4.0 agencies, but also for other problem-solving cases that the decision-making panel requires to advance their understanding of the matter at hand with the use of generative AI.
In spite of the significant contributions of this study, it is not free of limitations since it was a case study and the criteria generated are subjective. Given this study's other limitations in terms of expert panel composition, future research is recommended to incorporate more diverse panels to minimize potential biases. They can enhance the comprehensiveness of the decision outcome by inviting sustainability experts and marketing consultants, or end-user perspectives. Besides, in the proposed approach, ChatGPT served as a central component of the initial framework, condensing large amounts of data into organized decision criteria. Some studies (e.g. (Marcaccini et al., 2025; DEMİR et al., 2025)) support that ChatGPT, in comparison to other existing AIs, has demonstrated promising accuracy in the results it provides. However, it is recommended that future studies add to this study by using other AIs. Future research should look into how other large language model-based generative AI tools can compare the outcomes, validate and/or broaden the set of decision criteria. This study had a limited case company scope. Accordingly, the proposed approach is suggested to be tested and refined by further studies in different domains. It is advisable to carry out comparative research in different industries and geographical areas to enlighten our perception of the generalizability and compatibility of the proposed decision-making approach.
The author expresses gratitude to technology developers. The article benefited from AI technology through brainstorming, language editing and presentation clarity improvement.
Appendix Equations
Max Z.
Subject to:
