The finding is significant because food waste in the hospitality industry is not only an environmental problem but also a financial issue. Every kilogram of food discarded represents costs already incurred for purchasing ingredients, labor, energy, production, and waste handling.
This problem can become more complex in luxury hotels, where restaurants, buffets, and banquet services must maintain sufficient food availability despite fluctuating guest numbers. Producing too much food can result in unnecessary waste, while producing too little can affect service quality.
AI Provides Real-Time Food Waste Data
Before implementing AI technology, the hotel examined in the study did not have detailed information about the type, weight, and economic value of food waste. Kitchen waste was collected directly by a third party, leaving management with limited information about where the waste originated and how much it cost.
The introduction of an AI-based system changed this situation. The technology provided digital and real-time information about the type, quantity, and value of food waste generated during kitchen operations. Management could then use the information to evaluate food production, menu planning, purchasing, and food-cost control.
Initial data after AI implementation showed that fruit trimmings reached 351 kilograms, representing 23% of recorded waste, with a value of approximately Rp2.35 million. Plate waste reached 1,138 kilograms, or 76% of the recorded waste weight, while overproduction accounted for 8 kilograms with a value of approximately Rp108,917.
The availability of this detailed information allowed hotel management to identify specific sources of waste instead of relying only on total waste volume.
Five Stages of Food Waste Management
Widiantari, Rukmiyati, and Mareni examined AI-based food waste management using five aspects proposed by Karakas: Menu Planning, Purchasing & Storing, Food Preparation, Communication with Guest & Staff, and After Service.
In Menu Planning, AI-generated data helped identify food items that produced the largest amounts of waste. The information was used to evaluate menus, portion sizes, and production quantities.
One notable result was the reduction in overproduction. The value of overproduction declined from Rp108,917, equivalent to 8 kilograms, in February to Rp21,997, or 1.7 kilograms, in March. The hotel used guest-count forecasting and staged production, particularly for breakfast buffet items, to better match food production with actual demand.
For Purchasing & Storing, ingredient requirements were determined according to projected guest numbers. During the observation period, the hotel recorded zero stock-out incidents. AI-generated information about ingredients that frequently became waste was also used to evaluate purchasing and ingredient utilization.
In Food Preparation, fruit, vegetable, and meat trimmings were among the main sources of waste. In February, total trimmings reached 415 kilograms with a value of Rp3.28 million. The figure declined to 276 kilograms worth Rp2.11 million in March.
The Communication with Guest & Staff aspect involved guest feedback, employee training, operational briefings, and the use of the AI dashboard. However, the study also identified early resistance among some employees who considered the system an additional workload. Continuous training and coaching were therefore necessary to maintain consistent implementation.
At the After Service stage, food that still met quality and safety standards could be reused. Some surplus food was redirected for staff meals, sampling, rework, or subsequent service. Organic waste such as fruit trimmings was sent to third parties for composting or conversion into animal feed.
AI Contributes to Food-Cost Savings
One of the study's most important findings concerns the financial implications of AI-based food waste management. Between October 2025 and March 2026, estimated food-waste savings ranged from approximately Rp8.66 million to Rp12.30 million per month.
The estimated contribution to food cost was:
October 2025: 0.33%
November 2025: 0.44%
December 2025: 0.44%
January 2026: 0.39%
February 2026: 0.43%
March 2026: 0.39%.
From October 2025 through February 2026, the hotel's actual food cost remained below its standard food-cost level of 33.4%. In March 2026, however, actual food cost increased to 35.26%.
According to the study, the March increase was associated with a period of high tourist arrivals and operational demands surrounding Nyepi, Bali's Day of Silence. The hotel had to prepare larger quantities of food and offer promotions to accommodate increased guest demand. This demonstrates that food costs are influenced by factors beyond food waste alone.
Technology Alone Is Not Enough
Although AI provides detailed operational data, the study identified an important accounting limitation. Food waste was not recorded as a separate account in the hotel's financial statements. Instead, it was included within the Cost of Goods Sold (COGS).
As a result, the financial impact of food waste was not yet fully visible as a separate figure in the hotel's financial reporting. The researchers noted that AI-generated operational data alone cannot completely solve this issue without corresponding improvements in cost-accounting practices.
The study also highlights the importance of employees. Technology can identify waste, but employees still determine how consistently the system is used and how the information is translated into operational improvements. Training, discipline, communication, and organizational culture therefore remain essential components of successful food waste management.
Overall, the findings by Kadek Ari Widiantari, Ni Made Sri Rukmiyati, and Ni Ketut Mareni indicate that AI can serve as a strategic tool for luxury hotels seeking to reduce food waste and improve operational efficiency. By providing real-time information about the source, quantity, and economic value of food waste, AI enables hotel managers to make more data-driven decisions about purchasing, production, portion sizes, and surplus food management.
For hotels and restaurants in Bali and other tourism destinations, the approach offers a potential model for combining digital transformation, cost efficiency, and environmental sustainability. However, successful implementation requires not only investment in technology but also continuous employee training and accounting systems capable of making the financial impact of food waste more transparent.
Author Profile
Kadek Ari Widiantari is the corresponding author of the article and is affiliated with Politeknik Pariwisata Bali, Ministry of Tourism of the Republic of Indonesia. The article was co-authored by Ni Made Sri Rukmiyati and Ni Ketut Mareni, who are affiliated with the same institution. The source article does not provide the authors' academic degrees or specific fields of expertise.
0 Komentar