You are opening our English language website. You can keep reading or switch to other languages.

Travel Agency Solutions Proof of Concept: The NDC Chatbot

Travel Agency Solutions Proof of Concept: The NDC Chatbot

Challenge

New Distribution Capability (NDC) has led to complications in operations for travel agencies, travel management companies, online booking tools, and OTAs. Companies embarking on their NDC adoption journey require a streamlined and effective method to access the essential knowledge and expertise needed for their transition.
Location
Global

PoC Description

Travel Agency Solutions has developed an LLM-powered chatbot tool that facilitates the process of booking flights and assists with the selection of optimal travel options based on multiple sources of NDC data.

The chatbot provides users with a simple question-and-answer interface, offering flight options that match the specifications provided in the chat. The chatbot is also able to modify search results based on additional or alternative parameters, recalling earlier segments of the conversation and filtering data accordingly.

Beyond filtering results by price or date, the chatbot is able to navigate more complex requests, such as finding the most affordable flight within a specific month or only options operated by specific carriers and partnerships.

Business Benefits

  • The chatbot acts as a tech advisor, offering seamless flight offer guidance
  • Supports integration with specific and/or multiple aggregators, accommodating expanding offer options
  • Keeps an up-to-date record of what is currently supported within a company’s ecosystem
  • Assists internal stakeholders, engineers, and customers to understand how NDC may impact them

Process Flow

The intuitive chat interface of this PoC transforms the complex task of NDC navigation into a straightforward conversational experience.

1

User submits requests into chatbot based on required parameters

  •  
2

Requests processed through a web application in the chat

  •  
3

Model generates responses, which are shared with user through the interface

  •  
4

Seamless handling of follow-up questions within the same conversation

  •  
Example of JSON data for a flight request, sorted by price
Example of JSON data for a flight request, sorted by price
Example of conversation flow and data capture
Example of conversation flow and data capture

Technologies

LangChain
AWS Kendra
AWS ECS
AWS ECR
AWS Bedrock
Contact Us

Please provide your contact details and we will respond shortly.