Revolutionizing the hospitality industry: the impact of AI technologies on efficient food management and reduction of food waste Open Access

Food waste refers to the inappropriate use of food intended for human consumption, such as processing for animal feed or discarding edible food (Yusoff et al., 2024). The hospitality sector, which is a key part of the world’s tourism economy, contributes significantly to the economy by providing accommodation and food and beverage services to guests (Ampofo, 2020; Yadav et al., 2025). However, this sector faces a major challenge in the form of food waste, which can have serious environmental, social, and economic consequences. In catering facilities, which represent organizational units with the basic function of providing accommodation, food, and drinks to guests, 49% of the income is generated from the accommodation, 42% from food services, while the remaining 9% goes to the provision of other, supplementary services (Davis et al., 2018). Especially in large hotel chains, which serve thousands of meals every day in several outlets (restaurants, banquets, room service, etc.) a large percentage of food ends up not being consumed in the context and with the purpose with which it was transformed (Davis et al., 2018; Karakas, 2021). This waste includes leftover food from the professional kitchen of the catering establishment, unused food from the banquet, and unused food from the buffet tables and even in plate’s over portioning in classic restaurants (Elnasr et al., 2021; Hennchen, 2021). Food waste in the hospitality industry has both environmental and economic impacts. Environmentally, it contributes to methane emissions in landfills and results in the unnecessary use of resources involved in food production, transportation, and preparation (Bilska et al., 2022). Economically, it leads to significant losses for hotels due to the costs associated with purchasing, preparing, and storing uneaten food. By reducing food waste, hotels can lower operational costs and enhance profitability (Girotto et al., 2015; Schanes et al., 2018).

The problem of food waste in catering, which accounts for about 12% of the total amount of food waste, represents a serious challenge, but it has not been sufficiently investigated in academic studies (Eriksson et al., 2017; Tomaszewska et al., 2021; Wu et al., 2021; Tang et al., 2023). Although this issue is of great importance, there is a lack of comprehensive research in this area. The fight against food waste not only has significant environmental implications but also deeply affects the global economy (Shen et al., 2024). The agricultural sector, food production, and its distribution are key to many countries' economic stability and development (Chen et al., 2023; Rimhanen et al., 2023). Understanding food science and technology can lead to more efficient farming methods, improved production processes, and improvements in international food trade (Pawlak and Kołodziejczak, 2020). This progress is essential for economic development, job creation, and improved food safety standards (Carvalho, 2006; Doughan, 2020; Cabral et al., 2024).

This research aims to explore and assess the impact and potential application of artificial intelligence (AI) in reducing food waste within the hospitality industry, focusing on kitchen operations. The study seeks to provide innovative solutions to the current issue of food waste in hotels and restaurants by examining how AI can be integrated into various operational processes, such as inventory management, menu and dish planning, process automation, and real-time waste tracking. There are several key research gaps that this study aims to address. Firstly, while the issue of food waste in hospitality is widely recognized, previous studies have not comprehensively examined the role of AI in addressing this problem. There is a lack of research focusing on the specific applications of AI technologies in different aspects of hospitality operations, such as menu personalization and process automation. Additionally, few studies investigate the interaction between AI systems and hospitality employees, as well as their potential impact on efficiency and waste reduction. Notwithstanding, a delicate balance between the need for info registration and information overload is out there (Saxena and Lamest, 2018). The context of Serbia and Montenegro is particularly relevant, as both countries are in transitional phases regarding the adoption of sustainable practices in hospitality. Despite some efforts to introduce digital tools in food management, many establishments still rely on traditional methods. The infrastructural and regulatory limitations in waste tracking and recycling, along with limited investments in AI technologies, make the application of AI a significant challenge and opportunity in these regions.

While numerous studies have examined food waste management from operational and behavioral perspectives, theoretical integration of AI functionalities (e.g. process automation, real-time waste monitoring) within hospitality-specific frameworks remains underdeveloped (Chernyshev et al., 2023; Huy et al., 2023). Additionally, there is a lack of regionally contextualized research addressing how AI can be systematically embedded into hospitality practices in transitional economies like Serbia and Montenegro. This study contributes by proposing a comprehensive model that bridges this gap.

This study represents one of the first comprehensive attempts to explore the role of AI in reducing food waste in the hospitality sector, encompassing inventory management, menu and dish planning, process automation, and food recycling. The study’s innovation lies in the application of structural modeling for data analysis, which enables a deeper understanding of the interdependencies between various factors and their impact on operational efficiency. Notably, the study integrates technology, food management, and sustainability into a coherent model. The study contributes to the expanding literature on the application of AI in hospitality by providing new insights into the potential for reducing food waste. The findings can serve as a foundation for developing strategies that optimize operations in hotels and restaurants, reducing costs and increasing profitability while improving environmental performance. The study also lays the groundwork for future research on the long-term effects of AI technology implementation in the hospitality industry.

In this study, the term Artificial Intelligence (AI) refers to practical applications within hospitality operations (Huang et al., 2022; Kim et al., 2025), such as predictive analytics (machine learning), automated kitchen systems (robotics), and real-time monitoring tools (sensor-based data analysis) (Güngör and Yücel Güngör, 2024). Rather than focusing on a single AI subfield, the study encompasses a set of AI-enabled tools aimed at improving food management processes. This includes inventory forecasting algorithms, menu recommendation systems, cooking automation, and smart waste tracking solutions. Although this study does not evaluate specific AI algorithms or software, it focuses on how hospitality professionals perceive the role of AI-powered functionalities—such as automation, predictive analytics, and real-time tracking—in enhancing operational efficiency and reducing food waste.

The issue of food waste is becoming increasingly critical on a global scale, drawing attention from both political and social spheres (Schanes et al., 2018; Roka, 2022). This challenge encompasses not only discarded edible food but also the loss of essential resources like water, land, finances, and energy (Wunderlich and Martinez, 2018; Kibler et al., 2018). These losses have significant environmental and economic repercussions (Wunderlich and Martinez, 2018; Baig et al., 2019). Research on food waste has expanded, exploring various dimensions such as household waste, management strategies, sustainable practices, measurement techniques, and reduction approaches across the food supply chain (Petrović et al., 2017; Elimelech et al., 2018; Zrnić et al., 2023). According to the Food and Agriculture Organization, roughly one-third of global food production is either discarded or wasted unnecessarily (Bond et al., 2013; FAO, 2018). Food waste can be categorized into “food loss,” which covers uneaten food from unharvested crops to unused products in households and stores, and “food waste,” which includes raw or processed food that spoils during transport, fails to reach retail or catering operations, or is not consumed as intended (Alexander et al., 2013; Bilska et al., 2015, 2022). Reducing food waste is crucial for sustainability efforts, where it plays a central role (Woolley et al., 2021; Cappelletti et al., 2022; Gajić et al., 2024a).

In the United States, food waste from restaurants is a major concern, with these establishments producing between 9.98 and 14.97 billion kilograms of waste annually (Sakaguchi et al., 2018; Vardopoulos et al., 2025). Additional sectors, including schools, hotels, and hospitals, contribute another 3.18 to 4.99 billion kilograms each year (Sakaguchi et al., 2018; Dhir et al., 2020). Factors such as oversized portions, supply chain inefficiencies, and extensive menus lead restaurants to discard 4%–10% of purchased food (Evans and Nagele, 2018). Studies show that diners leave around 17% of their meals uneaten, and 55% of these leftovers end up as waste, exacerbated by issues like over-preparation and inadequate storage (Susskind and Chan, 2000). Buffet restaurants face particularly high waste levels due to food safety regulations preventing the reuse or donation of uneaten items. In developing countries, around 44% of global food loss happens during post-harvest and processing stages due to inadequate practices and poor infrastructure (Sakaguchi et al., 2018; Okumus et al., 2020). In contrast, developed countries, responsible for 56% of global food waste, see about 40% of this waste at the consumer level, driven by consumer habits, misunderstanding of expiration dates, and improper food handling. Additionally, the loss of traditional culinary knowledge, which helps reduce waste by utilizing leftovers, is another contributing factor (Karunasena et al., 2021; Quintero-Angel et al., 2022). Research from 2018 notes that within the European Union, 10% of the 88 million tons of food discarded annually is wasted (Sakaguchi et al., 2018; Caldeira et al., 2019; Okumus et al., 2020).

The “Zero Waste” concept represents a novel approach that encourages people to alter their lifestyles to minimize waste (Bogusz et al., 2021). In the context of Serbia and Montenegro, the issue of food waste remains underexplored despite its growing relevance. Serbia generates approximately 247,000 tons of edible food waste annually, with the hospitality sector contributing around 40,000 tons. A significant portion of this waste ends up in landfills, largely due to underdeveloped recycling infrastructure and the lack of food donation systems. In Montenegro, although precise data are scarce, the trends mirror those in Serbia, especially in tourist-heavy coastal regions where seasonal fluctuations lead to overproduction and increased waste. Furthermore, the hospitality industry in both countries faces challenges such as limited digitalization, insufficient staff training on sustainability practices, and a lack of government incentives for food recovery. These contextual factors underscore the importance of investigating AI as a potential solution for waste reduction in these specific regional settings (Chernyshev et al., 2023; Gajić et al., 2024a). Hospitality establishments, primarily restaurants, and hotels, are under great pressure to find ways to manage large amounts of food waste, which is caused by excessive portions, inefficient service methods, and a large menu offer (Öztürk et al., 2023; Filimonau et al., 2023). Priority in waste management includes prevention, reuse, recycling, and disposal. Some practices include composting and using waste to produce renewable energy (Tisi et al., 2023). Hotel waste management includes prevention, recycling, donation, composting and storage improvement, smart food merchandising, menu and dish design, staff training, and customer engagement (Filimonau and Delysia, 2019; Öztürk et al., 2023; Filimonau et al., 2023; Simões et al., 2023).

The integration of artificial intelligence in catering kitchens is becoming an increasingly important factor for improving efficiency in reducing food waste. With increasing pressures on sustainability in the hospitality sector, hospitality establishments, primarily restaurants, and hotels, are looking for innovative ways to optimize their operations, and AI technologies provide significant opportunities in this regard (Malefors et al., 2022; Onyeaka et al., 2023). One of the key areas in which AI contributes to improving efficiency is inventory management. Through the use of predictive analytics tools, restaurants can more accurately predict the demand for certain foods, thereby avoiding excessive procurement and, ultimately, food waste (Minaam et al., 2018). These systems can analyze historical data and take into account factors such as seasonal food availability, weather, and even media trends to provide more accurate forecasts (Minaam et al., 2018; Liegeard and Manning, 2020; Onyeaka et al., 2023).

Also, AI is used to automate the food preparation process, which reduces the time required to prepare meals and minimizes errors that lead to waste. Robotic arms and cooking monitoring systems can precisely manage the cooking process, maintaining consistent quality and reducing the need for chef involvement (Kumar et al., 2021; Taneja et al., 2023). In addition, AI technologies play an important role in menu planning and service personalization (Berezina et al., 2019). The application of machine learning algorithms allows restaurants to adapt the offer according to guest preferences and trends, which reduces the risk of wasting food that is not popular with them (Malefors et al., 2022). This personalization also improves guest satisfaction and increases their loyalty, which contributes to the restaurant’s long-term success (Bowden and Dagger, 2011; Berezina et al., 2019; Malefors et al., 2022). Another important aspect is the real-time monitoring of waste (Joshi et al., 2022; Vujić et al., 2022). In today’s world, where sustainability has become a key component of all aspects of business, the integration of artificial intelligence (AI) into inventory management in professional kitchens is an innovation that can significantly impact food waste reduction (Sinha and Praveen, 2024). AI technologies, such as machine learning algorithms and data analytics, provide the ability to accurately track and forecast ingredient needs in the kitchen. Such systems can predict needs based on historical data, consumption trends, and seasonal variations, thereby reducing the likelihood of over-ordering and unnecessary food storage (Addanki et al., 2022; Taneja et al., 2023).

AI systems enable restaurateurs to monitor the amount of food wasted and analyze the reasons for food waste. This information can be used to introduce corrective measures and optimize the food preparation process. In addition, this data can be used to educate employees and raise awareness about the importance of waste reduction in all stages (Martin-Rios et al., 2018; Joshi et al., 2022).

By applying advanced AI algorithms and process automation, restaurateurs can optimize their operations, reduce costs, and contribute to global sustainability efforts at the same time (Milton, 2024). Although this process is still under development, research and examples from practice so far show promising results. The study is grounded in the Resource-Based View (RBV), which posits that organizations achieve sustained competitive advantage through the strategic deployment of valuable, rare, inimitable, and non-substitutable resources. In this context, AI technologies serve as strategic assets that can enhance internal processes, reduce inefficiencies, and drive innovation. However, despite the growing attention to AI in operations management, little empirical research has examined its specific influence on food waste management in hospitality settings. This study extends RBV by demonstrating how AI functions as an enabling capability for sustainability-oriented performance in hospitality environments. Artificial intelligence technology represents an important step forward in the transformation of the hospitality sector towards a more efficient and environmentally conscious business (Blešić et al., 2014; Pirani and Arafat, 2016; Milton, 2024). Therefore, the initial, null hypothesis is defined as:

By applying AI to inventory management, professional kitchens can improve estimation accuracy and optimize food storage, which directly impacts waste reduction. AI systems can analyze consumption and inventory data in real-time, defining order points and drawing just-in-time supply strategies, enabling rapid order adjustments and balancing of grocery stocks. This results in a reduction in waste resulting from spoiled or expired food (Sachani et al., 2021; Taneja et al., 2023). The integration of AI technologies also enables guidance toward the right food-saving strategies, such as precise meal and dish planning and better portion management. For example, if the system notices a tendency for inadequate portions, it can suggest corrections in recipe cards (Boru et al., 2019; Helo and Hao, 2022).

AI can support the development of innovations in the cooking process, thereby finding new techniques that reduce waste, helping Chefs in this desire (Kumar et al., 2021; Cheah et al., 2022). Effective application of AI in inventory management enables better control over food quality (e.g. freshness of raw materials), which is of crucial importance in professional kitchens. In this way, not only the amount of waste is reduced, but also the quality of service and guest satisfaction is increased. AI systems can provide inspiration for new techniques and recipes, which can help maximize the use of available resources (Noone and Coulter, 2012; Addanki et al., 2022). In terms of financial aspects, integrating AI into inventory management can lead to significant savings. Reducing food waste reduces procurement and maintenance costs, which has the effect of reducing overall kitchen operating costs (Bisoi et al., 2020). With all these advantages, it is emphasized that the application of AI technologies represents a significant step forward in resource management and sustainability in professional kitchens (Noone and Coulter, 2012; Striebig et al., 2019; Camarena, 2020). Accordingly, the first hypothesis was defined, which reads:

While prior studies have addressed AI in general business or supply chain contexts (Lv et al., 2022; Milton, 2024), there is a lack of integrated theoretical frameworks explaining its role in food waste reduction specifically within hospitality kitchens. The application of AI in menu and dish planning and personalization of service in restaurants and hotels represents a significant advance in the field of food management and waste reduction (Leung and Loo, 2022; Lv et al., 2022; Milton, 2024). Analysis allows restaurants and hotels to adjust their menu and recipe cards based on these insights, thereby reducing the likelihood of over-ordering and thus potentially reducing food waste. For example, if the system notices that certain dishes are not delivering well or at all, it can suggest changes in the menu or adjust quantities for future production (Godfray et al., 2010; Howard, 2021; Ding et al., 2023). Service personalization, based on guest data analysis, can improve experience and efficiency. AI systems can gather information about guests' previous visits, preferences, and allergies, thereby enabling the creation of personalized menus that better suit each guest’s individual needs and preferences. This type of personalization not only increases guest satisfaction but can also reduce waste caused by inappropriate or undesirable menu options (Piccoli et al., 2017; Howard, 2021).

Additionally, AI can be used to predict future needs based on past data and seasonal food availability. These systems can provide recommendations on what foods to order in what quantities, based on past trends and expected changes in demand. In this way, restaurants and hotels can plan their supplies (Piccoli et al., 2017; Das, 2023). Reducing waste and optimizing inventory management directly impact operational cost savings. With all these advantages, the application of AI in menu planning and personalization of service can significantly improve the efficiency of restaurant and hotel operations while contributing to sustainability and reducing food waste (Leaven et al., 2017; Piccoli et al., 2017; Das, 2023). Accordingly, the second hypothesis was defined, which reads:

AI for automated monitoring of food preparation processes in professional kitchens is an innovation that can drastically improve efficiency and reduce operating costs. Using advanced algorithms and sensors, AI enables monitoring of all aspects of the preparation process, from the initial preparation of the food to the serving of the dish, that is, to the final preparation of the dish (Sachani et al., 2021; Addanki et al., 2022). Automated monitoring with the help of an AI system enables accurate monitoring of cooking time, temperature, and other critical parameters. These systems can continuously control the processes, ensuring that every step in food preparation is carried out according to the intended standards (Gajić et al., 2024c). This precision helps avoid errors and unnecessary deviations, which improves the quality of the dish and reduces the need for corrections (Sachani et al., 2021; Howard, 2021; Lutz and Coradi, 2022).

AI can analyze data from the preparation process and identify areas for improvement in efficiency. Based on the analysis, the systems can suggest the optimization of workflows, a change in the schedule of tasks, or a change in technical procedures, which can lead to savings in time and resources (Waltersmann et al., 2021; Vukolić et al., 2023). This type of analytics allows kitchens to better manage their operations and reduce inefficiencies. The integration of AI into the food preparation process also leads to improvements in quality management. AI systems can provide detailed reports and analysis on the quality of ingredients as well as preparation processes, enabling the rapid identification and correction of issues that may affect the quality of the dish (Sharma et al., 2021; Waltersmann et al., 2021). All these advantages contribute to reducing operating costs and increasing the overall efficiency of kitchen operations and restaurant revenue. With AI, professional kitchens can achieve higher standards of quality and productivity while reducing unnecessary costs and waste (Hennchen, 2019). Based on that, the third hypothesis was defined, which reads:

The integration of AI to track food waste in real-time in hotel and restaurant kitchens represents a significant advance in waste management. This technology enables the detection and analysis of food wastage in all stages of the preparation process, which is of crucial importance for the application of rapid corrective measures and the reduction of the amount of waste (Bibi et al., 2017; Lv et al., 2022). AI systems that monitor food waste in real-time use advanced sensors, cameras, and algorithms to continuously monitor and analyze consumption and waste data. These systems can identify moments when food is not being used optimally or when the estimated amount needed to prepare a meal is exceeded. The systems can immediately alert staff to any irregularities, allowing for a quick response (Maiyar et al., 2023). By applying AI to food waste monitoring, kitchens can get detailed reports on the types and amounts of waste. This analysis provides important insights into the causes of wastage, such as incorrect estimations of ingredient amounts, unforeseen losses during preparation, or premature preparation leading to expiration dates (Silvennoinen et al., 2019). Based on this data, restaurants and hotels can develop strategies to reduce waste, such as more precise menu planning or portion optimization (Liegeard and Manning, 2020; Maiyar et al., 2023).

AI systems can also help optimize storage and ingredient management. Real-time monitoring enables recognition of when ingredients are nearing the end of their shelf life or when they have been inadequately stored. This information can be used to adjust orders and storage, thereby reducing the risk of wastage due to spoiled food (Corradini, 2018; Chen et al., 2023). In addition, AI technology in this sense can help in training staff and raising awareness for food management and ultimately food waste (Martin-Rios et al., 2018). Analyzes and reports from the AI system can serve as a basis for developing training programs that will help employees understand the importance and techniques of waste minimization, thereby contributing to sustainability in the kitchen (Gajić et al., 2024b). Financially, integrating AI into real-time food waste tracking can provide significant savings. Reducing waste leads to a reduction in procurement and waste management costs, which improves the overall profitability and sustainability of the business (Azevedo et al., 2012; Martin-Rios et al., 2018). Consequently, the fourth hypothesis of this research was defined, which reads:

Applying AI to analyze and recycle food scraps in hotel and restaurant kitchens can significantly contribute to the development of more sustainable practices and the reduction of overall waste. This technology enables innovative approaches to food waste management, improving efficiency and sustainability in the hospitality industry (Ciccullo et al., 2022; Areche et al., 2022). AI systems can analyze food residue data, such as quantities, types, and causes of residue generation. With the help of algorithms, the systems can identify patterns and trends in the data, which enables a better understanding of the causes of waste and thus the development of strategies to minimize waste (Bibi et al., 2017; Abdallah et al., 2020). This analysis can show, for example, which processes or menu options lead to the most waste, enabling changes in food production and preparation (Matzembacher et al., 2020). One of the key aspects of applying AI in food waste recycling is the ability to optimize the recycling process. AI systems can manage technologies to process and convert food scraps into new resources, such as organic fertilizers or new products (Cheah et al., 2022). These systems can automatically control and direct food waste to the appropriate recycling processes, thus ensuring maximum use of all resources and reducing the amount of waste that ends up in landfills. The integration of AI into the recycling process also enables advanced methods for monitoring and managing the quality of recycled products (Arvanitoyannis et al., 2008; Wen et al., 2018). AI systems can continuously monitor conditions in recycling units, such as temperature and humidity, and adjust processes to ensure the best possible result. This type of control improves recycling efficiency and reduces the risk of contamination and failed processes (Ye et al., 2020; Cheah et al., 2022).

Additionally, AI can help educate and raise employee awareness about the importance of recycling. Food residue analyzes and reports can be used as part of employee training programs, helping them understand the importance of proper waste management and the role they play in the recycling process. The financial aspect of applying AI in food waste recycling is also significant (Liu et al., 2023). Reducing waste and optimizing recycling processes can lead to savings in waste management costs and improve business profitability. In addition, a positive impact on the environment and adherence to environmental standards can improve the image of a hotel or restaurant and attract environmentally conscious guests (Onyeaka et al., 2023). The application of AI for the analysis and recycling of food scraps represents an innovative step towards a more sustainable management of resources in hotel and restaurant kitchens, reducing overall waste and contributing to environmental protection (Ye et al., 2020; Onyeaka et al., 2023). In this regard, the fifth and last hypothesis of this research was defined, which reads:

According to the previously defined hypotheses, and similar research by Sinthiya et al. (2022), Liu et al. (2023), and Onyeaka et al. (2023), a research model was constructed with modifications (Figure 1):

To collect data, a meticulously designed structured questionnaire was developed to examine various facets of management related to food waste and the implementation of artificial intelligence (AI) in the hospitality industry. As no suitable existing instrument was found in the literature—which primarily focused on conceptual models, literature reviews, and general proposals for AI use and food waste reduction, the authors created a new questionnaire grounded in prior empirical and theoretical studies.

The instrument was specifically constructed for this study and is based on an extensive review of existing literature addressing AI applications in hospitality operations and food waste management. It was designed to measure five core constructs: AI-enabled inventory management, menu planning and personalization, process automation, real-time waste tracking, and food recycling. Each of these constructs was operationalized through 3 to 5 items, formulated as declarative statements. Responses were captured using a five-point Likert scale ranging from “Strongly disagree” (1) to “Strongly agree” (5), with a neutral midpoint score of 3.

As no validated instrument existed that comprehensively encompassed these AI functionalities in the context of professional kitchens, the questionnaire items were developed by adapting relevant terminology and concepts from established research in AI, sustainability, and hospitality operations. Content validity was ensured through expert review by three scholars specializing in hospitality management and AI application in service industries.

A preliminary pilot test was conducted with 15 respondents employed in the hospitality sector to evaluate the clarity of the items and assess internal consistency. Based on their feedback, minor revisions were made to enhance the wording and structure of the questions. Subsequently, a second pilot study was carried out with a sample of 30 participants whose profiles matched the characteristics of the study’s target population. The objective of this pilot phase was to confirm the comprehensibility and relevance of the questionnaire, as well as to conduct preliminary reliability testing. Necessary adjustments were made based on the results.

The study employed a purposive sampling method, as the target respondents were specifically managers and chefs within hospitality establishments who possess direct operational knowledge of food waste management and AI implementation. This approach was deemed appropriate given the study’s objective to examine practical applications of AI in professional kitchen environments. Due to the specific inclusion criteria, such as professional role and experience, random sampling was not feasible. The researchers approached potential participants directly in the field, prioritizing individuals who were both available and qualified to provide reliable insights. This method ensured the relevance and contextual accuracy of the collected data while enabling efficient data collection across a diverse range of hospitality settings in Serbia and Montenegro.

The reliability of the instrument was assessed using Cronbach’s alpha, which yielded a coefficient of 0.846, indicating satisfactory internal consistency. The final version of the questionnaire included 25 items and was administered in the field through direct, face-to-face interactions with managers and chefs in hospitality establishments. This method allowed researchers to provide immediate clarification of any uncertainties, though such interventions were minimal and did not influence the respondents’ answers or the study’s outcomes.

To ensure the ethical integrity of the research, participants were fully informed about the purpose of the study and gave their consent to participate. Anonymity and confidentiality were maintained throughout the data collection process in accordance with standard ethical protocols. The data collection period extended from August 2023 to March 2024.

All questionnaire items were closed-ended to allow for standardized data collection and easier statistical analysis. This format also ensured the comparability of responses across participants. In addition to the main constructs, the questionnaire included closed-ended items related to the respondents’ sociodemographic characteristics, such as age, gender, and education level. The use of closed questions and a consistent Likert scale enhanced the objectivity of the measurements and minimized the risk of subjective or moral bias. This structure contributed to the precision of the dataset and facilitated the reliability and validity of the subsequent statistical analysis.

The research included a sample of 234 participants from 117 catering establishments in the Republic of Serbia and the Republic of Montenegro. In the Republic of Serbia, 128 managers and chefs from 64 restaurants and hotels, distributed in city, mountain, and spa centers, participated. In the Republic of Montenegro, 106 managers and chefs were surveyed in 53 restaurants and hotels, which were located in urban and coastal areas (Table 1). Research in two different countries enables the collection of data from different social, economic, and cultural contexts. This indicates more diverse and complex results that may be more important and relevant for comprehensive scientific conclusions. Apart from that, the results from both countries are combined because people of similar origins and similar habits live in both countries, and according to the latest data, most tourists in Montenegro come from Serbia, and vice versa. Transnational research provides a broader sample that may better represent different demographic and social groups. This can help avoid bias that can arise if research is conducted in only one setting (see Figure 2).

To ensure an adequate sample size for reliable statistical analysis, the G*Power software was used (Kang, 2021). This software allows for precise determination of the required number of participants based on several key parameters. In this study, the primary parameters used were effect size (f² = 0.15), significance level (α = 0.05), and desired test power (1-β = 0.80). An effect size of 0.15 was selected as a medium effect size, which is common in social research where moderate effects between variables are expected. Based on these parameters, the G*Power software calculated that a minimum sample size of 150 participants was required to achieve the desired test power of 0.80 at a significance level of 0.05. This number of participants ensures that the study has sufficient statistical power to detect significant effects between the analyzed variables, thereby reducing the risk of a Type II error, which refers to the failure to detect a real effect. A sample size of 150 participants allows for precise hypothesis testing in this study, and the additional participants do not change the conclusions of the study but rather increase the reliability of the results. Given the geographical scope of the research, the sample included participants from both Serbia and Montenegro. Depending on the proportion of participants from each country, the sample provided representativeness and enabled the results to be generalized to the target population.

Table 2 shows the sociodemographic structure of the respondents.

According to the gender structure, the majority of respondents are men (62.8%, n = 147), while women make up 37.2% (n = 87) of the sample. These data indicate a slight male predominance in the sample. The geographical distribution of respondents shows that the participants are from two countries: Serbia (54.7%, n = 128) and Montenegro (45.3%, n = 106), which indicates a relatively balanced sample from both countries. As for jobs, the sample is almost evenly divided between managers (50.9%, n = 119) and chefs (49.1%, n = 115), which allows an even insight into the perspectives of both professional groups. When it comes to the educational profile of the respondents, most of them have completed university education (41.0%, n = 96), while a smaller percentage have secondary school (8.1%, n = 19), high school (25.2%, n = 59), master’s studies (24.4%, n = 57), and only 1.3% (n = 3) of respondents have doctoral education. This educational profile indicates that the majority of respondents have a higher education, which may be relevant for the analysis of competenciesand attitudes towards the subject of research. A high SD value for the level of education (0.924) indicates greater variations in the educational level of the respondents.

The gathered data was analyzed using basic descriptive statistics, specifically by calculating the average values and the variation from the mean, as indicated by the standard deviation. These statistical measures offered a clear understanding of the main patterns and the degree of variation in participants' opinions. In this research, a Cronbach’s alpha score of 0.891 was achieved, reflecting a strong internal consistency and high reliability in the responses gathered. Factor analysis, as shown in Table 7, was used to uncover underlying variables or constructs within the questionnaire data. This analysis revealed significant patterns and connections among the survey items, which improved the understanding of the critical factors influencing business practices in professional kitchens. Additionally, the Kaiser-Meyer-Olkin (KMO) measure yielded a value of 0.844, indicating that the dataset is well-suited for conducting factor analysis. Bartlett’s test yielded a chi-square value of 711.334 with 106 degrees of freedom and a significance level of 0.000. This statistically significant result (p-value < 0.05) indicates a strong correlation among the variables, suggesting that the observed relationships are not due to random chance, thereby confirming the data’s suitability for further factor analysis. Advanced statistical techniques, such as regression analysis and structural equation modeling (SEM) using SmartPLS, were employed to examine the relationships among variables and to test the proposed hypotheses. These methods offered crucial insights into the intricate interactions within the hospitality sector. To ensure both reliability and validity of the data and outcomes, several evaluation criteria were applied, including Cronbach’s alpha, composite reliability, average variance extracted (AVE), the Fornell-Larcker criterion, and the Standardized Root Mean Residual (SRMR). The model’s predictive capability for economic stability was indicated by an R² value of 0.481, suggesting that approximately 50% of the variance in economic stability is accounted for by the independent variables in the study. Additional confidence values will be detailed in the tables within the results section. Table 3 specifically highlights the reliability and validity metrics for all research constructs. Upon reviewing these metrics, it was found that while the constructs varied in their internal consistency, they generally exhibited sufficient reliability. All constructs in the model were specified as reflective measurement models. This decision was based on the assumption that the observed indicators are manifestations of the underlying latent variables, and changes in the construct would be reflected in all associated indicators. Accordingly, reliability and validity were assessed using standard criteria for reflective models, including Cronbach’s alpha, composite reliability, and average variance extracted (AVE).

Table 3 provides insight into the reliability and validity of various constructs within the research. The Cronbach’s Alpha values for most constructs demonstrate good internal consistency, with particularly high values for scales such as “AI Menu Planning and Personalization” (0.821) and “AI Inventory Management” (0.712), indicating the reliability of these scales. Scales like “AI Process Automation” (0.767) and “Efficiency and Waste Reduction” (0.762) also show solid results, while lower, but acceptable values were recorded for “AI Food Recycling” (0.702) and “AI Real-Time Waste Monitoring” (0.741). The rho_A values, which represent an alternative measure of reliability, confirm similar consistency as Cronbach’s Alpha, with minor variations that are within the expected limits. These values further affirm the reliability of the constructs. The Composite Reliability values for all constructs are above the threshold of 0.70, which is a key indicator of good consistency among the indicators. Particularly noteworthy are the high values for “AI Process Automation” (0.865) and “AI Inventory Management” (0.800), which indicate a strong correlation among the indicators within these constructs. The Average Variance Extracted (AVE) values, all above 0.60, show that the constructs successfully explain the variance of their indicators. This is particularly evident for “AI Menu Planning and Personalization” (0.833) and “Efficiency and Waste Reduction” (0.816), where the high AVE values indicate that the indicators well represent the measured construct.

In Table 4, as part of the structural equation modeling analysis using SmartPLS, the Fornell-Larcker criterion is presented to assess discriminant validity.

The values on the diagonal, which represent the square root of the average variance extracted (AVE), are greater than the correlation values between the constructs. This indicates that all constructs in the model have adequate discriminant validity, which means that they are sufficiently different and that each of them measures a different aspect of the use of artificial intelligence in the hospitality industry (Ab Hamid et al., 2017). Correlations between constructs are mostly moderate to low, which further confirms that the constructs are well-defined and that there is a clear distinction between different aspects. Although there are more correlations between individual constructs, such as “AI Process Automation” and “AI Food Recycling” and “Efficiency and Waste Reduction” and “AI Food Recycling,” those relationships still retain discriminant validity. These results suggest that the constructs are well conceived and that there is a clear distinction between different aspects of the use of artificial intelligence, which contributes to the reliability and validity of the overall research.

Table 5 presents the composite reliability values for each construct, which all exceed the commonly accepted threshold of 0.7, indicating that the measures used are reliable. The table also includes associated t-statistics, which are notably high, and p-values at the 0.000 level. These statistics confirm that the construct reliabilities are statistically significant, reinforcing the robustness and consistency of the constructs within the model.

Table 5 presents the results of the statistical analysis for various constructs in the research of the impact of artificial intelligence on the efficiency and reduction of waste in the hospitality industry of the Republic of Serbia and the Republic of Montenegro. Parameters in the table include original sample (O), sample mean (M), standard deviation (STDEV), t-statistics (|O/STDEV|), and p-values. The AI Inventory Management construct has an original sample (O) of 0.545, indicating a moderate relationship with the dependent variable. A t-statistic of 3.234 and a p-value of 0.006 indicate that this relationship is statistically significant, meaning that AI inventory management is a significant factor in the research. AI Menu Planning and Personalization with the original sample of 0.892, has the strongest relationship with the dependent variable. The t-statistic value is 6.000 and the p-value is 0.000, indicating a highly significant relationship. This suggests that menu planning and personalization through AI is an extremely important factor in process optimization. AI Process Automation has an original sample of 0.982 and it also shows a very strong relationship, with a t-statistic of 3.150 and a p-value of 0.007. The results indicate a significant contribution of process automation in the hospitality industry within this model. AI Real-Time Waste Monitoring is a construct that has a sample size of 0.659, indicating a moderate association, but a t-statistic value of 4.679 and a p-value of 0.000 indicate that this factor is statistically highly significant. This means that real-time waste tracking through AI has an important role to play in reducing waste. The AI Food Recycling sample construct of 0.787 also has a significant association with the dependent variable, a t-statistic of 5.054 and a p-value of 0.000, indicating that AI food recycling plays an important role in the waste reduction process. AI Integration is a construct that has a slightly lower value of the original sample (0.607) and t-statistic (2.050), a p-value of 0.017, and these data show that this factor is still statistically significant. This suggests that the integration of AI solutions is a relevant aspect of process optimization. Efficiency and Waste Reduction has an original sample of 0.605 indicating a moderate association, and a t-statistic of 2.234 and a p-value of 0.016 and these results show that this factor is statistically significant. It is emphasized that the efficiency and reduction of waste is significantly related to the introduction of artificial intelligence in catering processes.

The following section presents an in-depth analysis and interpretation of the research findings on the application of artificial intelligence in the hospitality industry of Serbia and Montenegro. It is organized into three key sections: the first part presents the results of the descriptive analysis, the second part discusses the findings from the factor analysis, and the third part covers the outcomes from the structural modeling. Table 6 details the mean values and standard deviations for each item, alongside the Cronbach’s alpha coefficients, which indicate the reliability of the measures.

The first conclusion from Table 6 is that there is significant potential for the use of AI for inventory management and waste reduction in professional hotel and restaurant kitchens. Statements with high mean scores, e.g. M = 4.58 for “It often happens that in professional kitchens there is an excess of supplies that eventually become waste” indicate the existing problem and the need for more efficient management of resources. Another conclusion is that the integration of AI technologies in various aspects of kitchen process management, e.g. “AI can help better personalized menus” with M = 4.15 visible. This aspect highlights the importance of applying AI in supporting hospitality establishments in improving service quality and performance. It can be noted that the variability in the results is reflected, where some claims show less support for AI in reducing waste such as “AI technologies that analyze guest preferences” with M = 3.09. This data points to the need for additional research and development of AI applications in these specific domains. The inclusion of AI in professional kitchens not only enables more efficient resources’ management but also emphasizes the importance of the sustainability and performance of catering establishments under modern technological progress. Low values of α (such as 0.801, and 0.811) indicate that the results were analyzed with significance and that the results are considered statistically significant if the data are close to these values. High values of α (e.g. 0.980, 0.961) indicate greater variability or the need for a larger volume of data to confirm the significance of the results.

Exploratory factor analysis with Promax rotation identified seven distinct factors, which are detailed in Table 7.

In the initial analysis, the component “AI Inventory Management” has the highest initial eigenvalue of 5.971, which explains 15.714% of the total variance, and indicates that this component has a significant impact on the model. As the components are ordered, the eigenvalues decrease, with each subsequent component explaining a smaller percentage of the variance. Other components, such as “AI Menu Planning and Personalization” and “AI Process Automation,” explain an additional 7.885% and 7.579% of the variance, respectively, leading to a cumulative percentage of explained variance of 31.178% after the three components. When looking at the rotated sums of squared loadings, it is observed that the rotated structure improves the interpretability of the factors, with each component retaining a significant portion of the explained variance. For example, “AI Menu Planning and Personalization” has a rotated factor loading of 4.243, which means that the rotation improved the explanation of the variance for this component. The cumulative percentage of the explained variance is 62.615% after rotating the seven components, which is a satisfactory result since more than half of the total variance of the data is explained by these factors.

Table 8 shows the results of the regression analysis. These results are crucial in evaluating hypotheses and in determining whether there is a significant relationship between the various variables in the model.

Table 8 shows the results of structural modeling for the impact of artificial intelligence on integration into different aspects of management. Each relationship between variables was analyzed through regression coefficients, standard deviations, t-statistics, and p-values. The first path, AI Inventory Management → AI Integration, has a coefficient of 0.649 with a standard deviation of 0.298, resulting in a t-statistic of 7.290 and a p-value of 0.026. This suggests that there is a statistically significant positive impact of AI-assisted inventory management on the integration of AI into systems. The second path, AI Integration → Efficiency and Waste Reduction, shows a very high coefficient of 0.897 with a standard deviation of 0.604, resulting in a t-statistic of 16.920 and a p-value of 0.000. This indicates a very significant impact of AI integration on increasing efficiency and reducing waste. The third path, AI Menu Planning, and Personalization → AI Integration, has a coefficient of 0.819 and a standard deviation of 0.261, with a t-statistic of 12.040 and a p-value of 0.039. This connection is significant, showing that menu planning and personalization through AI contribute significantly to the integration of AI. The fourth path, AI Process Automation → AI Integration, with a coefficient of 0.643 and a standard deviation of 0.347, has a t-statistic of 5.490 and a p-value of 0.000, indicating a very significant impact of process automation on AI integration. The fifth path, AI Real-Time Waste Monitoring → AI Integration, shows a coefficient of 0.717 and a standard deviation of 0.399, with a t-statistic of 4.299 and a p-value of 0.000. This result suggests a significant impact of real-time waste monitoring on AI integration. The last path, AI Food Recycling → AI Integration, has a coefficient of 0.634 and a standard deviation of 0.127, with a t-statistic of 6.971 and a p-value of 0.027. This shows the significant impact of AI-assisted food recycling on its integration. Structural modeling results show statistically significant impacts, with most p-values indicating high reliability of the results. These findings contribute to the understanding of how different applications of AI can improve integration and efficiency in different aspects of management.

Figure 3 shows the path coefficient model of the SmartPLS analysis results, focusing on the impact of various factors on economic stability.

The presented graph represents a structural model with latent variables and indicators. The model contains multiple latent constructs, and factors (represented by circles) such as: “AI Inventory Management,” “AI Food Recycling,” “AI Menu Planning and Personalization,” “AI Process Automation,” “AI Real-Time Waste Monitoring,” “AI Integration,” and “Efficiency and Waste Reduction.” Each of these latent constructs contains several indicators (represented by rectangles labeled IM, FR, MPP, PA, RTWM, EWR, and AI). All indicators have associated loading coefficients, which indicate how well individual indicators are related to their latent constructs. For example, indicators IM2 (0.958) and IM1 (0.884) have high values, suggesting that they are very significant for the latent construct “AI Inventory Management.” Similar analyzes can be performed for other latent constructs. The structural model also shows the association between different latent constructs. For example, there is a strong correlation between “AI Integration” and “Efficiency and Waste Reduction” (coefficient 0.897), indicating that the integration of AI technologies is key to increasing efficiency and reducing waste. Other relations between latent constructs have somewhat lower coefficients, such as the relations between “AI Integration” and “AI Food Recycling” (0.634) and “AI Inventory Management” (0.649), which shows that these factors are also significant but to lesser extent. measures. It should be noted that the loading coefficients for the indicators in the “AI Real-Time Waste Monitoring” construct are high, with RTWM1 (0.960) and RTWM5 (0.970) having the highest values. This indicates that these indicators are particularly important for measuring this latent variable.

Overall, the model shows the complex interdependencies between different aspects of the use of artificial intelligence in the hospitality industry, with a focus on resource optimization and waste reduction. Clear identification of key indicators and links between different constructive dimensions allows a better understanding of the factors that contribute to greater efficiency and less waste in this sector. Therefore, the results of this research point out that all the hypotheses were confirmed, which indicates the importance of this research.

This study provides a comprehensive insight into the role of artificial intelligence (AI) in reducing food waste and improving operational efficiency in the hospitality sector, with a particular focus on kitchen operations. Our results demonstrate that integrating AI technologies into key aspects of business, such as inventory management, menu planning, process automation, and real-time waste tracking, can significantly contribute to waste reduction. This waste reduction is directly reflected in lower operating costs, improved sustainability, and increased profitability of catering facilities. The findings of this research are consistent with previous studies that emphasize the importance of applying AI technologies to enhance operational efficiency. Bilska et al. (2022) demonstrated that AI can improve accuracy in inventory management, leading to a reduction in oversupply and, consequently, food waste. These results were further confirmed by Malefors et al. (2022), who highlighted that AI technologies play a key role in accurately predicting ingredient needs, thereby enabling inventory optimization and minimizing waste. Our study builds on these findings by providing additional evidence of how AI can enhance menu personalization, which not only reduces the amount of uneaten food but also increases guest satisfaction.

Furthermore, the automation of kitchen processes through AI technologies has proven to be crucial in reducing operating costs and increasing precision in food preparation. These findings align with the research of Sachani et al. (2021), who emphasized that automation can reduce errors in food preparation, lower operating costs, and improve service quality. Our study further validates these claims, demonstrating that AI can significantly contribute to precision and efficiency in kitchen operations, directly leading to a reduction in food waste. This aspect is particularly important because reducing errors and optimizing operations result in significant savings in both resources and labor costs. Another important aspect of our study is the role of AI in real-time waste monitoring. Joshi et al. (2022) emphasized the importance of such waste tracking systems in hotels and restaurants, and our results confirm this assertion. Specifically, our findings show that AI enables quick reactions to irregularities in the food preparation process, thereby reducing the amount of waste. This ability to monitor and adjust in real-time represents a significant step forward in efficient resource management and sustainable hospitality operations.

The ecological implications of our results are also noteworthy. Our study supports the findings of Shen et al. (2024), who demonstrated that reducing food waste has profound implications not only for reducing greenhouse gas emissions but also for conserving resources. The application of AI technologies enables the optimization of resources in a manner that directly contributes to reducing the ecological footprint of the hospitality sector. This is particularly important in the context of global efforts to preserve the environment and achieve sustainable development. These results are consistent with similar research, such as the works of Camarén (2020) and Cappelletti et al. (2022), who underscore the importance of AI technologies in the global context of waste reduction and resource conservation. Our study further validates these claims, showing that integrating AI into hospitality operations is key to achieving more sustainable practices and long-term economic efficiency. Moreover, our research indicates that AI not only brings economic benefits but also significantly contributes to environmental preservation, an increasingly important aspect of the modern hospitality business.

The results of this study have direct and meaningful implications for the hospitality industry, particularly in regions such as Serbia and Montenegro, where resource constraints and infrastructural limitations challenge operational efficiency. Industry professionals can harness AI functionalities not only to reduce food waste but also to improve cost control, optimize kitchen workflows, and enhance the overall guest experience. Specifically, predictive analytics for demand forecasting, machine learning algorithms for personalized menu planning, and sensor-based technologies for real-time waste monitoring represent the most relevant AI specializations for the sector. Integrating these systems into daily operations requires not just technological investment but also staff training, digital culture development, and alignment with broader sustainability goals.

Moreover, AI technologies should not be viewed as isolated tools but rather as embedded, adaptive systems that support data-driven decision-making at every level of kitchen and inventory management. This is especially crucial in small and medium-sized enterprises, where automation and precision can substitute for labor shortages or inefficiencies. The findings also suggest that AI applications in food recycling—though less developed—have the potential to contribute to circular economy practices in hospitality, if further supported by infrastructure and education.

From a theoretical standpoint, this research offers a novel contribution by synthesizing various AI-driven functionalities into a unified framework that explains their collective impact on operational efficiency and sustainability. The model presented in this study expands existing literature by demonstrating that AI can serve as a strategic resource within the Resource-Based View (RBV) framework, especially in transitional economies. This not only validates previous assumptions about the benefits of AI but also contextualizes them within specific regional and organizational realities, thereby offering a more nuanced understanding of technological adoption in hospitality.

The study was carried out in selected cities across Serbia and Montenegro, focusing on hotels and restaurants in both mountainous areas and other regions. This geographical limitation could affect how representative the sample is, as not all regions of these countries were included. While the findings are relevant to the specific areas studied, generalizing these results to the entire countries should be done with care. It is also important to recognize that job structures in the hospitality industry vary depending on the location. Not every establishment follows a uniform hierarchy with managers and head chefs at every level. Moreover, these structures can differ based on the size and organizational setup of each establishment. This variability in job roles within the hotel sector can influence how business operations and responsibilities are managed. Thus, when evaluating the research and planning future actions, it’s crucial to consider these differences and tailor strategies to fit the unique organizational models and business frameworks of various hospitality establishments in Serbia and Montenegro.

This study presents important theoretical contributions in multiple areas. Firstly, it enhances the current body of knowledge on how artificial intelligence affects food waste in the hospitality industry, offering fresh insights into these complex relationships. This addition to the literature is vital for deepening the theoretical comprehension of the issue. Moreover, the research employs a multidisciplinary approach by combining elements such as artificial intelligence, catering, and food management. This comprehensive perspective aids in synthesizing various disciplines, providing a more nuanced understanding of their interplay. The study’s findings are not only theoretically significant but also have practical applications. They offer actionable insights for creating strategies to minimize food waste in the hospitality sector, effectively linking theoretical concepts with real-world practices. By addressing food waste from multiple angles, this research underscores the importance of this issue, potentially increasing awareness and prioritization among hospitality leaders, including managers and chefs. Additionally, the study establishes a theoretical foundation for future research, highlighting key factors that affect the influence of artificial intelligence on food waste in the hospitality sector, which could inspire further exploration in this and related fields.

The findings of this research emphasize the crucial role of artificial intelligence in minimizing food waste within the hospitality industry. Restaurants and hotels are encouraged to adopt AI-driven strategies, such as improved inventory management, staff training, and partnerships with organizations focused on food redistribution. Implementing these measures can lead to reduced operational costs and higher profitability. Increasing awareness of the consequences of food waste is also essential, as it can drive significant change. Collaboration between hotels, restaurants, consumers, and other stakeholders is vital for educating the public and promoting sustainable, responsible practices. The practical takeaways from this study underscore the importance of integrating artificial intelligence to reduce food waste, which is key to ensuring the stability, sustainability, and social responsibility of hospitality businesses. A cooperative effort among all involved parties is recommended to achieve meaningful progress in this area. The practical implications of this research indicate the importance of the role of artificial intelligence (AI) in improving efficiency and reducing waste in the hospitality sector. Research has shown that various aspects of applying artificial intelligence, such as inventory management, personalized menu planning, process automation, and real-time waste monitoring, significantly contribute to reducing waste and increasing operational efficiency. The results indicate that the integration of artificial intelligence into various operational processes is key to optimizing inventory and reducing food losses. Inventory management with the help of artificial intelligence enables better forecasting of needs and avoidance of over-ordering, which directly contributes to waste reduction.

Personalized menu planning, which uses artificial intelligence, enables better adaptation of the food offered to the needs and preferences of guests, which results in a reduction of waste due to unsold dishes. This implication may be particularly useful in the context of hospitality establishments that face large variations in demand. Automation of processes in the hospitality industry, through the use of artificial intelligence, significantly reduces human errors and improves accuracy and efficiency in performing daily activities. This contributes to the reduction of waste, as errors in the preparation and serving of food are reduced. Real-time waste tracking systems, which use artificial intelligence, allow management to spot key points where waste is generated and take corrective action in real-time. This ability significantly improves resource management and reduces the negative impact on the environment. Finally, this paper also shows that the constructs associated with the application of artificial intelligence are well defined and have a significant impact on efficiency and waste reduction, which provides a basis for further development of technologies and strategies in this direction. Hospitality establishments that adopt these technologies can expect improved operational efficiency, reduced costs, and better compliance with environmental standards, which will position them as leaders in an industry that is increasingly striving for sustainability.

Although this study offers important insights into the application of artificial intelligence in hospitality food management, it also opens several avenues for future research. First, the study was limited to the context of Serbia and Montenegro, which may affect the generalizability of the results. Future studies could examine similar models in other geographical or cultural contexts to validate and compare the findings. Second, while this study focused on five specific AI functionalities (inventory management, menu planning, process automation, waste monitoring, and recycling), future research could explore additional AI applications such as customer interaction, supply chain optimization, or energy efficiency in kitchen environments. Third, further qualitative research involving interviews with hospitality managers and kitchen staff could provide deeper insights into barriers and facilitators of AI adoption. Longitudinal studies may also be useful to observe the effects of AI integration over time. Lastly, future work could refine the measurement of sustainability outcomes, not only in terms of food waste reduction but also with regard to carbon footprint, water usage, and broader environmental indicators.

This study provides valuable insights into the role of artificial intelligence (AI) in modern hospitality, where it addresses the growing challenges related to sustainability and operational efficiency. The analysis of various business aspects, including inventory management, menu planning, process automation, and real-time waste tracking, has demonstrated that the application of AI can significantly reduce food waste and enhance operational efficiency. In a world where resources are increasingly limited and environmental demands are rising, the integration of AI technologies into hospitality operations represents a crucial step towards more sustainable and profitable business practices. Research based on empirical data collected from various hospitality establishments, not only confirms the importance of AI in cost reduction and resource optimization but also opens new perspectives for the application of advanced technologies in everyday business operations. The findings of this study provide solid evidence of how AI can transform traditional resource management methods and contribute to greater sustainability in the hospitality sector. Through this analysis, the research has shown that AI not only facilitates everyday operations but also equips managers with tools for making more informed decisions, which can have long-term positive effects on business performance. The introduction of AI technologies into inventory management and menu planning can significantly reduce waste, while process automation and real-time waste tracking enable quicker adaptation to changes and increase operational efficiency. These results clearly demonstrate that AI has the potential to become an essential tool for every modern hospitality establishment striving for sustainability and competitiveness in the market. At a time when resource pressures and environmental standards are at their highest, the integration of these technologies into business practices is not only desirable but also necessary for the long-term success and development of the hospitality sector.