Artificial Intelligence-Driven Supply Chain Agility and Resilience: Pathways to Competitive Advantage in the Hotel Industry

Abstract

Background: The extraordinary disturbances faced by the hotel industry, ranging from worldwide health problems to political instability and climate change, have highlighted the insistent need for more resilient and agile supply chain (SC) systems. This study explored how artificial intelligence (AI) capabilities can generate competitive advantage (CA) through supply chain agility (SCA) and supply chain resilience (SCR) as mediators and competitive pressure (CP) as a moderator. Methods: Drawing on the resource-based view (RBV) framework, we suggested and empirically tested the study model. Using data collected from 432 hotel managers and analyzed using Partial Least Squares Structural Equation Modelling (SEM-PLS). Results: the results reveal that AI-driven SC can significantly strengthen SCA and SCR. Furthermore, SCA and SCR can act as powerful mediators, and CP can strengthen the tested relationships (the links from AI adoption and CA) as a moderator. Conclusions: The study made several theoretical and practical contributions by integrating AI capabilities into SCR and SCA frameworks in the hotel and tourism context, and by providing practical evidence for professionals aiming to leverage AI-driven SC tools to navigate uncertainty and create sustainable CA.

1. Introduction

The hospitality sector is still affected by major technological changes, as the industry continues to grow and global travel demand rises [1]. Hotels operate in a highly competitive environment characterized by volatile demand, high customer expectations, and increasing supply chain disruptions, making it more difficult to achieve a sustainable competitive advantage. The authors of [2,3] identified two main strategies for gaining a competitive advantage: differentiation and cost leadership. Nowadays, technological innovation is a crucial factor in maintaining a long-term differentiated competitive advantage [4]. Recently, supply chain innovation has emerged as a key competitive factor enabling hotels to improve performance within a turbulent business environment [5].

The concept of the hotel supply chain has emerged as a significant focus in the service management literature [6]. However, most existing research emphasizes the advantages hotels gain from effective supply chain management and its positive influence on hotel performance [7]. Given that hotel supply chains, as a subset of service supply chains, face significant challenges in synchronizing operations and sustaining performance amid increasing uncertainty, volatility, and dynamism in the contemporary business environment [8], it is important to recognize that the literature on hotel supply chain management has yet to sufficiently address the role of artificial intelligence (AI) or explore strategies and practices that leverage AI to foster supply chain innovation and capabilities, such as agility and resilience. These capabilities are essential for enhancing supply chain performance and providing hotels with a distinct competitive advantage [5].

Tourism supply chains (TSCs), including hotel SCs that are one of the most important components of it, are the result of businesses forming connections with a wide range of industry stakeholders, including suppliers, distributors, competitors, governments, and other businesses, due to the cross-cutting, interdependent, and fragmented nature of tourist products [9]. Moreover, a wide range of participants in both the public and private sectors are involved in the tourism supply chain (TSC), which is defined as “a network of tourism organizations engaged in various activities ranging from the supply of different components of tourism products/services like flights and accommodation to the distribution and marketing of the final tourism product at a specific tourism destination” [10]. The complex nature of the tourism and hospitality industry makes it even more susceptible to failure when supply chain resiliency is low or nonexistent [11].

The complex, uncertain, vulnerable, and unstable business landscape that is being created by the acceleration of change and the frequency and magnitude of unanticipated and unfavorable events (such as pandemics, wars, disasters, and political and social unrest) can result in high levels of uncertainty, stress, and anxiety at the individual, organizational, and societal levels [12]. Under these circumstances, forecasting hotel occupancy rates is a revenue management method closely related to the sustainability of the hotel industry [13]. Therefore, in the “new normal”, resilience and agility have frequently been presented as a panacea for ensuring the supply chain process succeeds in both domestic and international tourism [12]. According to [14], the hotel’s ability to employ digital technology to generate and deliver value to customers is the only solution nowadays. Additionally, according to the “Dynamic Capability View” (DCV), resilience approaches such as preparedness, response, and recovery are important for minimizing supply chain risks and fostering higher performance in the tourism industry [15]. In this context, artificial intelligence is a strategic tool capable of reshaping hotel supply chains by improving demand forecasting, inventory management, enhancing coordination with suppliers, and supporting real-time decision-making [16].

Accordingly, this study adopts a Dynamic Capability View to address existing research gaps by examining how AI-driven technologies might enhance supply chain agility (SCA) and supply chain resilience (SCR), providing hotels with sustainable competitive advantages (CA). Competitive pressure (CP) is used as a moderating variable in the study to shed light on how external market dynamics affect the extent to which AI-driven capabilities improve supply chain resilience and agility, thereby influencing the organization’s capacity to maintain a competitive edge in the hospitality and tourism sector.

2. Theoretical Background

2.1. Artificial Intelligence (AI) and Competitive Advantage (CA)

Businesses in the hospitality sector benefit from technology investments because they increase competitive performance and differentiate themselves in global marketplaces [17]. This phenomenon can be explained by the resource-based view (RBV), which posits that an organization’s key resources determine how well it performs [18,19]. According to this hypothesis, rare, valuable, unique, and inimitable resources can provide an organization with a competitive advantage by improving business performance and adding value [18,19]. Among these, artificial intelligence capabilities are becoming a more significant and intangible resource for enhancing business performance [20,21,22].

In the hospitality industry, AI and robots are now used across many areas, leveraging vast amounts of data, advanced algorithms, and powerful computing to perform service tasks in ways similar to humans [23]. AI is a revolutionary technology for generating novel services, making decisions, and shaping how businesses interact with final customers, and it employs machine learning techniques and smart computing in the workplace [24]. In this context, the hotel industry is continually under severe pressure to control costs and work more effectively. In response to these pressures, artificial intelligence of things (AIoT) technology has become vital in the current hospitality industry because it can deliver better service, better utilize valuable resources, and foster new ideas [25]. Additionally, it enables personalized guest experiences, data-driven demand forecasting, automated booking systems, and improved back-end operational efficiency [26].

The integration of artificial intelligence (AI), including machine learning, natural language processing, and generative AI, into supply chain management (SCM) enhances efficiency, resilience, and strategic decision-making. Thus, improving essential SCM processes, including customer relationship management, inventory management, transportation, procurement, demand forecasting, and risk management [27]. Additionally, AI analyses data in forecasting demand and pricing, whereas IoT enhances supply chain reliability, cost-efficiency, and agility by improving data exchange and information flow. Blockchain is the most widely adopted innovation in the tourism and hotel industry, facilitating sustainability, operational efficiency, and customer satisfaction [28]. Consequently, it suggests that artificial intelligence may give companies a competitive advantage [29]. Considering these considerations, the following hypothesis can be developed:

H1. 

The adoption of artificial intelligence (AI) has a positive direct effect on competitive advantage (CA) in the hotel industry.

2.2. Artificial Intelligence (AI) and Supply Chain Agility (SCA)

Supply chain agility is the ability to predict and respond promptly and effectively to irregular shifts in the business [30]. Additionally, the ability to adapt to the unpredictable dynamics of the business environment is referred to as supply chain agility [30]. The ability of organizations to instantly adjust their operations and act proactively is also a pattern of agile practices [13]. Adaptability in relationships between businesses and suppliers and partners, in supplying goods or services, and in meeting customer expectations is essential to agile supply chain functions [31]. Hotel business can promptly adapt to a continuously unexpected changing environment by leveraging data analytics and AI-based technology, such as in-time inventory management and selling forecasting [13]. This need for agility is particularly critical in tourism, where one important aspect is the uncertainty surrounding future demand [10]. This uncertainty stems from the fact that many stakeholders interact intricately to create tourism goods, which are seen as value-added chains of various service components [10]. It is possible to construct the following hypothesis from this discussion:

H2. 

The adoption of artificial intelligence (AI) has a positive direct effect on supply chain agility (SCA) within hotel supply chains.

2.3. Supply Chain Agility (SCA) and Competitive Advantage (CA)

Since the tourism and hospitality sector comprises a wide range of operationally distinct organizations, such as hotels, restaurants, and travel agencies [32], additionally, it comprises an integrated network of horizontal, vertical, and diagonal providers that form a tourism supply chain, aiming to offer travelers a complete range of tourism-related goods and services [33]. Therefore, service supply chain activities are considered inherently more complex than manufacturing [34]. In this context, tourism supply chains benefit from agility, which enables them to respond appropriately to changes in their dynamic environments [35]. To ensure consistent service delivery, supply chain companies in the tourism industry must focus on improving their agility [36]. Thus, agility serves as an essential dynamic capability that supports competitive advantage in uncertain environments [30]. Specifically, achieving this advantage requires leveraging the assets and capabilities of supply chain participants and stakeholders, both upstream and downstream [13]. Finally, businesses’ competitive advantages and flexibility have been demonstrated to increase with efficient supply chain management (SCM), particularly in dynamic marketplaces [37]. After this conversation, the following hypothesis is put forth:

H3. 

Supply chain agility (SCA) in hotel supply chains has a positive direct effect on hotels’ competitive advantage (CA).

2.4. Artificial Intelligence (AI) and Supply Chain Resilience (SCR)

In the context of supply chains, resilience is a complex concept that has gained increasing attention amid global disruptions such as pandemics and natural disasters [37,38]. It is conceptualized as “the ability of a system to return to its original state, within an adequate period, after being disturbed” [38]. Specifically, supply chain resilience (SCR) is explained as the capacity of organizations to sustain operations, adjust, and recover promptly from disturbances to safeguard business continuity in unexpected circumstances [39]. Previous evidence has suggested several key factors that can foster business resilience, such as AI digital transformation, network design, and strategic alliances [40]. As per [15], resilience approaches (i.e., awareness, response, and recovery practices) are vital for minimizing supply chain hazards and improving performance in the hotel industry. For instance, customer service process and inventory management can be improved through data analytics and automation, thereby fostering operational efficiency and service quality [41]. AI-based technologies can improve supply chain efficiency, enhance risk estimation, and strengthen businesses’ ability to manage uncertainty [42]. Specifically, AI can foster supply chain resilience by improving quality and enhancing supply-demand matching [43], improving operational stability [44], expanding forecasting accuracy, improving logistics, and successfully managing inventory levels [42]. Based on the conversation above, a hypothesis can be proposed:

H4. 

The adoption of artificial intelligence (AI) has a positive direct effect on supply chain resilience (SCR) in the hotel industry.

2.5. Supply Chain Resilience (SCR) and Competitive Advantage (CA)

Supply chain resilience enables hotels to continue operations, adapt, and bounce back quickly from setbacks, ensuring sustainability and continuity in unpredictable situations [40]. Resilience solutions, such as readiness, reaction, and recovery measures, are also needed to reduce supply chain risks and enhance performance [15]. Consequently, enterprise supply chain resilience has become a key determinant of sustainable competitiveness, surpassing conventional operational efficiency considerations [45]. The theoretical underpinning for this strategic importance is found in the dynamic capacity theory (DCT) and the knowledge-based view (KBV), resilience and robustness are essential bridging mechanisms that convert market knowledge into performance results by allowing businesses to adjust and remain stable in the face of disruptions [46]. Furthermore, a resilient supply chain can improve following a disruption rather than merely returning to the disrupted form [47]. Hence, in this discussion, the following hypothesis is formulated:

H5. 

Supply chain resilience (SCR) in hotel supply chains has a positive direct effect on hotels’ competitive advantage (CA).

2.6. Supply Chain Agility (SCA) and Supply Chain Resilience (SCR)

Agility enables firms to predict disruptions and respond effectively, fostering stability and recovery across supply chain [35]. Crucially, this agility enhances resilience through proactive sensing and rapid adaptation mechanisms [30]. In the dynamic tourism sector, this relationship is particularly important, as agility enables organizations to respond appropriately to environmental changes, thereby strengthening business resilience and ensuring the continuity of service delivery [12,35]. Consequently, the supply chain in the tourism industry needs to focus on building its resilience and agility to ensure stable service delivery [48]. This discussion leads us to the following hypothesis:

H6. 

Supply chain agility (SCA) in hotel supply chains has a positive direct effect on supply chain resilience (SCR).

2.7. Supply Chain Agility as Mediator

AI-based technologies can improve the agility of organizations’ supply chain functions by fostering flexibility and visibility [13]. Moreover, AI can facilitate data sharing and prompt responses to unexpected environmental changes, which, collectively, can help gain a stainable competitive advantage [37]. Accordingly, the hypothesis below can be introduced:

H7. 

Supply chain agility (SCA) mediates the relationship between AI adoption and competitive advantage (CA) in the hotel industry.

2.8. Supply Chain Resilience as a Mediator

The integration of AI-based technologies in the supply chain process can enable organizations to expect the surrounding disruptions, adjust quickly to unexpected market changes, and recover more effectively [37], through advancing data analytics and digital partnership [38]. In the same vein, improving supply chain resilience has become a vital concern for organizations aiming to sustain a competitive edge over time and to reduce external disruptions [42]. Therefore, AI can indirectly improve competitive advantage by fostering supply chain resilience [45].

H8. 

Supply chain resilience (SCR) mediates the relationship between AI adoption and competitive advantage (CA) in the hotel industry.

2.9. SCA and SCR Sequentially Mediate the Link from AI and CA

Agility can foster resilience by supporting preparedness and recovery capabilities [30]. In addition, agility and resilience are correlated dynamics that enable organizations to recover from external distress [35]. Within this context, AI-based technologies can improve agility and resilience by fostering data-driven responsiveness [13]. Consequently, supply chain resilience and agility can convey the effect of AI on competitive advantage [45]. Based on this, the following hypothesis is proposed:

H9. 

Supply chain agility (SCA) and supply chain resilience (SCR) sequentially mediate the relationship between AI adoption and competitive advantage (CA) in the hotel industry.

2.10. Competitive Pressure (CP) as a Moderator

Competitive pressure is the pressure on businesses in the same industry to preserve or strengthen their competitive advantage [49]. Moreover, one of the main drivers of changes in business behavior is competitive pressure [50]. To stay ahead, hotels in a highly competitive industry must continually introduce new ideas and implement competitive strategies that leverage innovative technology [51]. According to previous studies, when a company realizes that using technology will provide it with a competitive edge and eventually improve its performance, competitive pressure influences that adoption [52,53,54]. Indeed, economists generally agree that competition increases the likelihood that innovations will be adopted [55]. Moreover, competitive pressure accelerates the adoption of AI and big data analytics, increasing transparency and reducing information asymmetry, thereby further improving supply chain operations [56]. This dynamic is especially pronounced in the hospitality sector, where businesses feel pressured to adopt new AI technologies and policies as this results in reduced expenses, more effective procedures, and faster information flow [57]. Thus, the following hypotheses are formulated (and see Figure 1):

H10. 

Collaborative partnerships (CPs) with suppliers positively moderate the relationship between AI adoption and supply chain agility (SCA) in hotel supply chains.

H11. 

Collaborative partnerships (CPs) with suppliers positively moderate the relationship between AI adoption and supply chain resilience (SCR) in hotel supply chains.

3. Methods

3.1. Measures

All employed measurement scales were derived from well-established previous instruments. AI variable was assessed using 5 items [58]. Sample items included: “We possess the infrastructure and skilled resources to apply AI information processing system” and “We use AI techniques to forecast and predict environmental”. Competitive advantage (CA) was measured using a 7-items scale [59]. For instance, “Compared with our competitors, we offer unique benefits and novel features to our customers” and “Compared with our competitors, we offer high quality products to our customers”. Supply chain agility (SCA) was captured through a nine-item scale [60]. For example, “Speed in increasing levels of product customization” and “Speed in improving customer service”. Supply chain resilience (SCR) was assessed using 5 items adopted from the (Belhadi et al., 2024) [22] study, which were originally derived from the prior studies [61,62]. For example, “Our hotel’s supply chain is well prepared to face constraints of supply chain disruptions” and “Our hotel’s supply chain can swiftly return to its original state after being disrupted”. Lastly, competitive pressure (CP) was operationalized through 3 variables [63]. For instance, “Our hotel seeks AI-driven solutions from its suppliers because our competitors are also demanding similar AI solutions from their suppliers” and “Our industry is progressively shifting toward adopting AI-oriented production and business processes.” To ensure scale content validity, the measurement scale was pretested by 21 experts in the hotel industry (10 academics and 11 professional practitioners). The panel confirmed that all the questions of the employed scale are clear, adequate, and relevant. Hence, no substantive revisions were required.

3.2. Data Collection

We have targeted surveying five- and four-star large hotels in Egypt, as these types of luxury enterprises are prioritizing AI digital transformation, specifically the adoption of AI-based technologies. The data was collected with the assistance of the author’s colleagues, who are higher education postgraduate students enrolled in hotels and tourism schools to which the authors are affiliated and who are currently working in the field. Hotel managers were directly contacted and provided with the survey link and QR code to complete the questionnaire. The survey link and QR code was consequently shared with other department managers and supervisors. Middle- and top-level positions within hotels are the focus, as these roles are actively involved in developing and implementing innovative strategies. Additionally, individuals in these roles contribute significantly to both the formulation and execution of organizational strategies. Thus, they possess sufficient information relevant to the study’s focus [64]. An introductory section at the beginning of the developed questionnaire to explore the study’s main objectives and explain confidentiality concerns. Respondents acknowledged that answering the survey questions constituted informed consent. Furthermore, they were informed that the survey has no right or wrong answers. Between July and September, a total of 432 complete and usable responses were collected after three reminder messages were sent. The mandatory-response feature in the online survey link helped in facilitating the intended aim. Following established guidelines [65] for a 95% confidence level and a 5% margin of error, researchers typically use the next-highest category in the sample size table when the exact population size is unknown. For populations with over 100,000 individuals, a sample of 384 is considered adequate for robust statistical power and generalizability. This study’s sample of 432 respondents surpasses that benchmark, reinforcing the reliability and validity of its findings. The sample comprised 61.3% male and 38.7% female respondents. The majority were aged between 25 and 35 years (56.9%), followed by those aged 36 to 50 years (34.3%). By educational level, most participants held a bachelor’s degree (66.4%), while 18.3% reported having a postgraduate degree.

3.3. Data Analysis

Given that the proposed research model comprises one independent variable (artificial intelligence), two mediating variables (supply chain agility and supply chain resilience)—each functioning both as an individual mediator and as sequential mediators—a dependent variable (competitive advantage), and finally a moderating variable (competitive pressure) operating along two statistical paths, the model can be considered relatively complex. Moreover, as the primary aim is to predict relationships among the variables rather than to confirm an existing theoretical framework, the Partial Least Squares Structural Equation Modeling (PLS-SEM) technique was deemed appropriate and thus applied using SmartPLS software, version 3. The PLS-SEM analysis was conducted in two stages. The first stage involved assessing the measurement (outer) model for convergent validity (evaluated using Cronbach’s alpha, item loadings, composite reliability, and AVE) and discriminant validity (assessed using the Fornell–Larcker criterion and the HTMT) (see Appendix A). The second stage involved evaluating the structural (inner) model using R², Q², path coefficients (β), and t-values [66].

4. Results

4.1. Common Method Bias (CMB)

To test common method bias (CMB), several statistical procedures were conducted. First, “Harman’s single-factor” test was run, as advised by [67]. The test indicated that a single factor could explain only 40.23% of the overall variance. This value is below the 50% threshold, revealing that CMB was not a serious issue. In addition, diagnostic checks supported the robustness of the study data. All “variance inflation factors” (VIF) scores ranged from 1.518 to 2.958 (

Table 1. Measurement model findings.

Dimensions	λ	VIF	Mean	SD	SK	KU
1. AI: (α = 0.887, CR = 0.917, AVE = 0.689)
AI_1	0.829	2.296	3.657	1.317	−0.605	−0.791
AI_2	0.860	2.721	3.775	1.250	−0.665	−0.612
AI_3	0.858	2.573	3.762	1.316	−0.738	−0.662
AI_4	0.832	2.092	3.630	1.418	−0.625	−0.929
AI_5	0.769	1.856	3.451	1.327	−0.427	−0.925
2. CA: (α = 0.923, CR = 0.938, AVE = 0.685)
CA_1	0.837	2.615	3.428	1.417	−0.281	−1.257
CA_2	0.825	2.958	3.573	1.385	−0.380	−1.233
CA_3	0.835	2.859	3.692	1.307	−0.506	−0.932
CA_4	0.799	2.403	3.512	1.369	−0.428	−0.998
CA_5	0.828	2.643	3.625	1.343	−0.511	−0.945
CA_6	0.839	2.612	3.634	1.302	−0.554	−0.755
CA_7	0.831	2.419	3.833	1.394	−0.783	−0.806
3. SCA: (α = 0.916, CR = 0.932, AVE = 0.632)
SCA_1	0.783	2.201	3.199	1.261	−0.149	−0.892
SCA_2	0.746	2.187	3.019	1.285	0.110	−0.924
SCA_3	0.732	2.134	2.840	1.312	0.254	−0.931
SCA_4	0.792	2.341	3.053	1.354	0.072	−1.075
SCA_5	0.831	2.715	3.148	1.358	−0.097	−1.140
SCA_6	0.834	2.946	3.070	1.380	0.046	−1.178
SCA_7	0.801	2.550	2.910	1.451	0.135	−1.285
SCA_8	0.833	2.649	3.130	1.487	−0.071	−1.367
4. SCR: (α = 0.917, CR = 0.938, AVE = 0.751)
SCR_1	0.864	2.544	3.576	1.428	−0.546	−1.066
SCR_2	0.872	2.748	3.720	1.292	−0.655	−0.640
SCR_3	0.878	2.909	3.685	1.318	−0.644	−0.712
SCR_4	0.854	2.611	3.655	1.350	−0.633	−0.789
SCR_5	0.865	2.650	3.688	1.334	−0.628	−0.746
5. CP: (α = 0.775, CR = 0.869, AVE = 0.689)
CP_1	0.781	1.518	3.602	1.294	−0.520	−0.728
CP_2	0.863	1.704	3.685	1.237	−0.486	−0.738
CP_3	0.844	1.604	3.748	1.252	−0.626	−0.623

Note: SK = skewness, KU = kurtosis, λ = loadings, α = Cronbach’s alpha, CR = composite reliability, AVE = average variance extracted, VIF = variance inflation factor, SD = standard deviation.

), well below the suggested threshold of 5.0, indicating no multicollinearity among the predictors [68]. Finally, skewness and kurtosis values fell within adequate ranges (skewness: −0.783 to −0.254; kurtosis: −7 to 0.612; see Table 1), confirming the assumption of normality in the current data [69].

4.2. Measurement Model Assessment

Prior to examining the structural paths, the measurement model was evaluated. Because traditional model fit indices typically used in CB-SEM are either unavailable or deemed unsuitable for PLS-SEM, given its variance-based nature [64], alternative criteria for assessing reliability and validity were employed. Following the guidelines of Hair et al. [68], first, “convergent validity (CV) was examined using item loadings (λ), Cronbach’s alpha (α), composite reliability” (CR), and the “average variance extracted” (AVE). As depicted in Table 1, all factor loadings are above the threshold of 0.70 (ranged from 0.732 to 0.878), CR (ranging between 0.869, competitive pressure—CP, and 0.938, competitive advantage—CA) and α (varied from 0.775, competitive pressure—CP, and 0.923, competitive advantage—CA) values fulfilled the suggested minimum value of 0.70, and AVE score are above the 0.50 criterion, specifically, the AVE score are ranged from 0.632 (supply chain agility—SCA) to 0.751 (supply chain resilience—SCR). Accordingly, these results support an adequate CV.

DV was tested with the Fornell–Larcker metric, which imposes that the AVE square root of each single factor should exceed its correlations with all other factors [70]. As depicted in

Table 2. Fornell–Larcker values.

	AI	CA	CP	SCR	SCA
AI	0.830
CA	0.435	0.828
CP	0.459	0.436	0.830
SCR	0.542	0.558	0.532	0.867
SCA	0.476	0.540	0.436	0.616	0.795

, this criterion was met, declaring the distinctiveness of all factors.

The “Heterotrait–Monotrait” ratio (HTMT) was also inspected to assess discriminant validity. As suggested by Henseler et al. [71], an HTMT score of 0.90 indicates that factors are theoretically related, whereas an accurate threshold of 0.85 is preferred when factors are more distinctive. As depicted in

Table 3. HTMT values.

	AI	CA	CP	SCR	SCA
AI
CA	0.475
CP	0.535	0.505
SCR	0.592	0.601	0.625
SCA	0.521	0.583	0.507	0.668

, all HTMT scores are below 0.85, providing additional evidence for discriminant validity. Taken together, these findings indicate that the measurement model met the established standards of construct reliability and convergent and discriminant validity, supporting the robustness of the model for subsequent structural model analysis.

4.3. Structural Model and Testing Hypotheses

The model explained substantial portions of variance in CA (R² = 0.385), SCR (R² = 0.533), and SCA (R² = 0.309). The interpretation of R² values is highly context-dependent, and scholars have offered varying benchmarks. Ref. [72], for example, classifies R² values of 0.75, 0.50, and 0.25 as substantial, moderate, and weak, respectively. Conversely, Tavakol and Dennick [73] argue that in behavioral research, even an R² of 0.20 may be considered strong. In light of these perspectives, the present study’s R² values fall within or exceed the cited ranges, indicating that the proposed model offers an adequate level of explanatory and predictive power. Additionally, the predictive relevance, as indicated by Q² values, was well above zero [68]. Additionally, the goodness-of-fit (GoF) index was estimated using the formula proposed by Tenenhaus et al. [74], where GoF scores of 0.10, 0.25, and 0.36 correspond to small, medium, and large effect sizes, respectively.

$$\mathrm{GoF}=\sqrt{{AVE}_{avy}\times {R^2}_{avy}}$$

The GoF obtained a value of 0.531, which is significantly above the suggested threshold for a large effect size, signaling a convincing overall fit of the study model. Furthermore, the ongoing debate on PLS-SEM highlights the value of reporting the “standardized root means square residual” (SRMR) as an approximate measure of model goodness-of-fit. In the current study, the SRMR score was 0.054, which is below the suggested threshold of 0.08, indicating adequate model fit [75]. Additionally, the “Normed Fit Index” (NFI) was also expected to measure the overall model fit [75]. Previous evidence suggested that NFI values between 0.60 and 0.90 demonstrate an adequate level of fit [76,77]. In this study, the NFI score was 0.853, which lies comfortably within the suggested range, confirming the adequacy of the model fit.

After verifying the reliability and validity of the outer model measures and following Henseler’s [78] suggested steps, hypothesis significance was evaluated using t-values, standardized path coefficients (β), and p-values from a structural model estimation with 5000 bootstrapped resamples.

The structural model findings provide support for the hypothesized relationships (

Table 4. Hypotheses evaluation.

Hypothesis	β	t	p	F2		Conclusions
Direct effects
H1: AI → CA	0.130	2.149	0.032	0.019		✔
H2: AI → SCA	0.394	7.760	0.000	0.166		✔
H3: SCA → CA	0.287	5.049	0.000	0.079		✔
H4: AI → SCR	0.304	6.482	0.000	0.125		✔
H5: SCR → CA	0.311	4.975	0.000	0.085		✔
H6: SCA → SCR	0.357	10.205	0.000	0.188		✔
Single mediating effect				CI
H7: AI → SCA → CA	0.113	3.948	0.000	0.063	0.175	✔
H8: AI → SCR → CA	0.094	3.902	0.000	0.050	0.146	✔
Sequential mediating effect
H10: AI → SCA → SCR → AI	0.044	3.779	0.000	0.024	0.069	✔
Moderating effect
H11a: AI × CP → SCA	0.140	3.080	0.002	0.048	0.225	✔
H11b: AI × CP → SCR	0.161	4.370	0.000	0.085	0.227	✔
Competitive advantage		R2	0.385	Q2	0.244
Supply Chain Resilience		R2	0.533	Q2	0.368
Supply chain agility		R2	0.309	Q2	0.179

Note: AI = artificial intelligence, CA = competitive advantage, SCA = supply chain agility, SCR = supply chain resilience, CP = competitive pressure, ✔ = supported, β = path coefficients, CI = confidence intervals.

and Figure 2). Artificial intelligence (AI) demonstrated a significant positive effect on competitive advantage (CA), supply chain agility (SCA), and supply chain resilience (SCR), supporting H1, H2, and H4. Both SCA and SCR were also found to significantly enhance CA, confirming H3 and H5. Moreover, AI was shown to indirectly influence CA through both SCA (β = 0.113, t = 3.948, p = 0.001, CI [0.063–0.175]) and SCR (β = 0.094, t = 3.902, p = 0.001, CI [0.050–0.146]), with significant partial single mediation effects, as well as a sequential partial mediation path through SCA and SCR (β = 0.044, t = 3.779, p = 0.001, CI [0.024–0.069]), supporting H7, H8, and H9.

In addition, the moderating role of competitive pressure (CP) was also significant, amplifying the impact of AI on both SCA (β = 0.140, t = 3.080, p < 0.001) (Figure 3) and SCR (β = 0.161, t = 4.370, p < 0.001) (Figure 4), confirming H10 and H11. All latent variables were mean-centered before creating the interaction terms, in line with established recommendations to reduce multicollinearity [79].

5. Discussion and Implications

Artificial intelligence (AI) has significantly transformed the hospitality industry [80]. Within this sector, the supply chain is a specialized area where AI is increasingly us to enhance operational efficiency [81]. While extensive research has examined artificial intelligence (AI) technologies and information technology innovations in the hospitality sector, there is a notable lack of studies assessing perceptions of AI adoption in supply chain management [82]. This study addresses this gap by empirically investigating the impact of AI utilization in hotel supply chains on competitive advantage, considering the mediating roles of supply chain agility and supply chain resilience, and the moderating effect of competitive pressures.

The study’s findings revealed that artificial intelligence has a positive direct effect on competitive advantage. (H1). This finding is consistent with previous studies, such as those by [29], which argued that artificial intelligence (AI) could provide organizations with a competitive advantage. Further, it reduces workload and boosts productivity by serving as an intelligent assistant that helps workers handle repetitive and procedural tasks [23]. In the hotel context, the implementation of information technologies addresses deficiencies in hotel food supply chain management caused by inadequate information flows by enhancing transparency and flexibility [82,83]. Furthermore, studies have emphasized that adopting green technologies enhances the efficiency and capabilities of hotel supply chains, reduces costs, and strengthens hotels’ competitive performance [84,85].

Regarding (H2), the study’s findings showed that artificial intelligence has a positive, direct effect on supply chain agility. This result aligns with [86], who found that artificial intelligence technologies are being aggressively implemented by hospitality firms as essential corporate tasks to address a range of daily management challenges. Furthermore, AI provides cutting-edge methods to expedite numerous procedures, including scheduling housekeeping and inventory management [26]. Additionally, AI’s ability to manage and assess massive datasets enhances the flexibility of the hotel supply chain in pursuing sustainable development goals and enables more environmentally friendly operational decisions [87].

Regarding (H3), supply chain agility has a positive direct effect on competitive advantage. Agility is a vital dynamic trait that fosters competitive advantage in uncertain situations [30]. Further, it has been demonstrated that efficient supply chain management boosts a business’ adaptability and competitive advantages, particularly in dynamic marketplaces [37]. However, the management of the tourism supply chain confronts greater difficulties in sustaining performance and organizing its operations because it is a service supply chain [88].

The PLS-report revealed that AI can positively and directly impact supply chain resilience, supporting H4. AI improves the decision-making process in supply chain management [89] by increasing forecasting accuracy, streamlining transportation, and actively managing inventory levels. Furthermore, the study’s findings demonstrated that supply chain resilience can positively and directly impact competitive advantage, confirming H5. In the context of competitive advantage theories, enterprise supply chain resilience (as a valuable and unimitated source of competitiveness) has surpassed traditional tools of operational efficiency [45]. Developing SCR has also become vital for organizations aiming to reduce external, unexpected, and unseen disruptions and sustain over time [42].

The findings also demonstrated that SCA can directly impact SCR, supporting H6. Agility can help organizations visualize unseen problems and respond effectively [35]. When businesses become more agile, they can control unexpected changes instantly and act proactively [30]. The findings also displayed that SCA can mediate the link from AI to CA, supporting H7. In the supply chain, employing artificial intelligence makes companies more agile by fostering their ability to visualize what is happening, adapt to changes, and make better decisions [13]. Artificial intelligence also assists businesses in a competitive environment by enabling them to respond more quickly to unexpected changes through predictions and shared data [37].

The PLS-SEM findings further revealed that SCR can mediate the link from AI to CA, supporting H8. Artificial intelligence-based technology can significantly enhance resilience by improving predictive capabilities and accelerate recovery [42]. When AI is integrated into supply chain operations, organizations can also anticipate disruptions, adapt to changing market environments, and recover more quickly [37]. Therefore, by increasing supply chain resilience, AI indirectly improves competitive advantage [45].

The study’s findings revealed that supply chain agility and resilience sequentially mediate the relationship between artificial intelligence and competitive advantage (H9). Artificial intelligence helps organizations adjust and respond quickly by making it easier to use data and collaborate (agility) [13]. Being agile also helps organizations become more resilient by making them better prepared to bounce back [30]. The combination of supply chain agility and resilience demonstrate how AI helps companies gain a competitive advantage [45].

Consequently, enhancing agility and resilience in hotel supply chain management through the implementation of artificial intelligence technologies can reduce challenges related to uncertainty, improve organizational anticipation and rapid adaptation, and ultimately facilitate the provision of customer-oriented services that often surpass guests’ expectations.

Moreover, the results demonstrated that competitive pressure positively moderates the relationship between artificial intelligence and supply chain agility (H10). Competitive pressure can accelerate the adoption of AI to further improve supply chain operations, maximizing information transparency and minimizing information asymmetry [56]. Similarly, organizations are under pressure to adopt new AI systems and technologies faster than their competitors to reduce costs, implement more effective procedures, and accelerate information flow [57]. Finally, the study results demonstrated that competitive pressure can positively moderate the relationship between artificial intelligence and supply chain resilience, supporting H11. Supply chain management has been under growing pressure from rivals as a result of higher degrees of unpredictability and uncertainty in business settings [90]. Thus, businesses are driven to improve their supply chains in order to gain a competitive edge over their direct rivals [91]. To foster innovation, organizations may also be under pressure from competitors to improve their access to market data and innovation capabilities [48]. However, supply chain management and technology adoption are significantly affected by competitive pressure in the hotel industry [90].

Overall, the hotel supply chain consists of customers, suppliers, clients, and the hotel’s internal departments. This network delivers products and services from primary suppliers to end users, specifically hotel customers [92]. Optimal supply chain efficiency depends on implementing information technologies, such as artificial intelligence, which facilitate on-demand information exchange across all stages and can reduce costs by 2.2% to 12.1% [93]. The supply chain adopts an integrated approach that encompasses the planning and monitoring of materials, logistics, and information, forming a comprehensive process from suppliers through the hospitality establishments or service providers to the final customer. This approach signifies a fundamental shift in business practices, as firms now compete as integrated supply chains rather than as isolated entities [94]. The integration of artificial intelligence within the supply chain enhances flexibility and adaptability to changing conditions and competitive pressures, thereby supporting the achievement of competitive advantage.

Theoretically, this study improved our understanding of SC digitization and management by presenting a refined perspective on how AI can reshape organizations’ competitiveness through dynamic supply chain capabilities. A main theoretical contribution is that AI is not purely a technological instrument, but an enabler of resilience. While previous studies have mainly focused on AI’s operational efficiencies, such as error reduction and predictive accuracy, this paper provides empirical evidence that AI functions as a high-level dynamic capability by responding to and recovering from environmental disturbances more swiftly and accurately. The outcomes expand the existing theoretical foundations of dynamic capability theory, showing that resilience and agility are not independent operational attributes but rather mediating strategic mechanisms through which AI competences can be translated into a sustainable CA. This re-explains AI-derived supply chain capabilities not as isolated outcomes, but as a mechanism for the business to redesign processes, reallocate resources, and adapt to ongoing disturbances. This paper provided a comprehensive theoretical-based explanation of the mediating roles of SCR and SCA in the mutual key. Resilience might serve as an adaptive instrument, translating AI analytical outcomes into agile decision-making and prompting the reconfiguration of SC activities. Conversely, resilience emerged as a preventative tool, leveraging AI-boosted insights, early detection, and strengthening supply chains against disturbances. These mediating roles demonstrated that AI could operate as a tool for exploiting unseen opportunities (resilience) and justifying risks (agility). This duality calls for a more nuanced theoretical framework in the SC literature, one that transcends linear or single-function approaches.

In practice, this paper offered several recommendations for businesses aiming to create a competitive advantage through AI-driven SC. The results supported the idea that AI can directly create CA by fostering functional efficiency and improving decision-making. This underscores the need for companies to increase investment in advanced analytics platforms, predictive algorithms, and real-time data integration tools. Decision-makers should prioritize AI systems that predict demand, improve inventory management, and enable smart procurement. Specifically, the implementation of artificial intelligence for demand forecasting in hotels facilitates more effective inventory control and reduces food waste in hotel food supply chains.

Second, the findings highlight the mediating role of supply chain resilience, underscoring the critical importance of operational flexibility. Companies should adopt AI-powered planning software to rapidly analyze potential scenarios, dynamically guide operations, and adaptively schedule production. These systems enable the supply chain to respond quickly to shifts in demand, market trends, or disruptions. In practice, organizations should integrate AI into their sales and operations planning processes.

Third, given AI’s powerful impact on supply chain resilience, organizations are advised to leverage AI to quickly identify risks and pinpoint deviations. Additionally, since hotel supply chains are designed to achieve customer satisfaction, artificial intelligence can be leveraged to link supply chain processes to real-time customer data and preferences. This integration enhances hotels’ ability to personalize their logistics operations (such as tailored in-room amenities and dynamic menu adjustments), thereby exceeding guest expectations and improving overall satisfaction. Fourth, it should be recognized that the strategic value of AI increases significantly under competitive pressure. The adoption of AI technologies should be accelerated. Management should treat AI implementation as a competitive imperative. Fifth, the study provides a methodological framework for resource allocation and capacity development. To improve the impact of artificial intelligence, priority should be given to employee training, skills development programs, cross-functional digital transformation teams, and investment in data governance.

6. Limitations and Avenues for Future Research

Despite providing valuable insights into the key role of AI in improving SCA and SCR in the hotel and tourism industry, this paper has several limitations that should be addressed. First, the paper relies mainly on cross-sectional survey data, which may limit its ability to capture the long-term impacts of SCA and SCR on CA. A longitudinal study design could offer a deeper understanding [95] of how AI-driven capabilities have developed over time and how businesses can translate them into sustainable CA. Second, this study uses self-reported survey data, which may introduce subjective bias despite efforts to ensure reliability. Future research should use objective measures or archival data to validate these findings. Third, the data were collected primarily from employees in the hotel sector, which might limit the generalizability of the study’s results to broader contexts. Future research is encouraged to collect data from wider samples across different industries. Fourth, the conceptual model focuses on SCA and SCR as mediators and CP as a moderator, but other business factors, such as leadership style and employee AI background, might also impact the tested relationships. Finally, the rapidly evolving nature of AI-driven technologies poses several ongoing challenges. Emerging AI technologies, such as generative large language models (LLMs) and blockchain–AI hybrids, might initiate new mechanisms that impact the supply chain. Future research papers should explore how the next generation of AI systems can reshape supply chain strategies and whether they can generate sustainable CA.