Amazon Lex Airline Solution

This solution deploys a pre-built Amazon Lex bot for airline travel assistance, providing conversational experiences for both voice and chat modalities.

Solution Overview

The Amazon Lex Airline Solution creates a complete conversational bot that helps users with airline-related tasks such as booking flights, checking flight status, and managing reservations. The solution includes:

• Amazon Lex bot with pre-built conversation flows
• Lambda functions for business logic and fulfillment
• DynamoDB table for storing sample data
• Optional Amazon Connect integration

Architecture

The solution consists of the following components:

1. Amazon Lex Bot: Provides the natural language understanding and conversation management
2. Lambda Functions:
   • Business Logic Function: Handles the core functionality and fulfillment
   • Lex Import Function: Imports the pre-built Lex bot configuration
   • DynamoDB Import Function: Populates the DynamoDB table with sample data
   • Connect Import Function: Sets up Amazon Connect integration (optional)
3. DynamoDB Table: Stores sample customer and flight data
4. IAM Roles: Provides necessary permissions for the components to interact

Deployment Options

CloudFormation Deployment

aws cloudformation create-stack \
  --stack-name AirlineSolution \
  --template-body file://amazon_lex_airline_solution_fixed.yaml \
  --parameters \
    ParameterKey=BotName,ParameterValue=AirlinesBot \
    ParameterKey=BusinessLogicFunctionName,ParameterValue=AirlinesBusinessLogic \
    ParameterKey=DynamoDBTableName,ParameterValue=Airlines_db \
  --capabilities CAPABILITY_IAM

Bash Script Deployment

./deploy_airline_solution_fixed.sh

Parameters

Parameter	Description	Default
BotName	Name of the Lex bot	AirlinesBot
BusinessLogicFunctionName	Name of the Lambda function for validation and fulfillment	AirlinesBusinessLogic
DynamoDBTableName	Name of the DynamoDB table for sample data	Airlines_db
ContactFlowName	Name of the Connect contact flow (optional)	AirlinesContactFlow
ConnectInstanceARN	ARN of Connect Instance (optional)	-

Testing the Bot

Using AWS CLI

aws lexv2-runtime recognize-text \
  --bot-id YOUR_BOT_ID \
  --bot-alias-id YOUR_BOT_ALIAS_ID \
  --locale-id en_US \
  --session-id "test-session" \
  --text "I want to book a flight"

Using Python Script

import boto3
import uuid

# Configure these values
BOT_ID = "YOUR_BOT_ID"
BOT_ALIAS_ID = "YOUR_BOT_ALIAS_ID"

# Initialize Lex client
lex_client = boto3.client('lexv2-runtime')
session_id = str(uuid.uuid4())

# Send a message
response = lex_client.recognize_text(
    botId=BOT_ID,
    botAliasId=BOT_ALIAS_ID,
    localeId='en_US',
    sessionId=session_id,
    text="I want to book a flight"
)

# Print the response
print(response)

Sample Utterances

• "Hello"
• "I want to book a flight"
• "Check my flight status"
• "What's the status of my flight?"
• "I need to change my reservation"

Cleaning Up

To delete all resources created by this solution:

# If deployed with CloudFormation
aws cloudformation delete-stack --stack-name AirlineSolution

# If deployed with bash script
./delete_airline_resources.sh

Resources

• Sample Customer Data
• Amazon Lex Documentation
• AWS Lambda Documentation
• Amazon DynamoDB Documentation

License

This solution is licensed under the MIT-0 License. See the LICENSE file.