Year: 2026
Type: dissertation
Abstract: This thesis presents the design and development of an AI-based automation system for managing restaurant customer communication using the Model Context Protocol (MCP). The study addresses a specific operational challenge in the hospitality industry, manual email management. Initial market research conducted for this study revealed that restaurant representatives spend hours daily processing multiple customer inquiries, creating a clear demand for efficient automation solutions. Consequently, the aim of this study was to automate general restaurant tasks and answer general inquiries through natural language processing. The research followed a design and implementation approach in which a prototype system was developed and evaluated through simulated customer email scenarios. The proposed solution is the combination of an AI model with backend services to create a seamless workflow for handling customer interactions. The MCP plays a key role in ensuring that there is a structured context which allows the system to coordinate between AI models and database operations for booking and cancellation processes. The system was evaluated using multiple test scenarios which simulate the emails of customers in the real world. The results demonstrate that the accuracy of intent detection and information extraction was high when there is well structured input. However, some limitations were noted in the management of ambiguous language and multi-intent requests. In addition, the research also examines the cost effectiveness and efficiency of the system. The result indicates that there was a considerable reduction in response time and human workload compared to manual processing. Overall, the research shows that integrating MCP with AI based automation provides a scalable and efficient solution for managing customer communication in restaurant booking systems.
Source: Theseus (Ammattikorkeakoulujen)
Authors: Dhakal, Prabin, Shrestha, Asmita
Institution: Helsinki Metropolia University of Applied Sciences
Language: English
Topic: AI in Service Interactions
Subfield: Marketing
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7162266578