Business Needs

Many businesses admit that chat technology is more efficient for engaging people than email, phone, or native app. This technology is being applied to various domains. Through our research, most people feel more comfortable to interact with Chatbot and it never makes informative or emotional mistakes. This may lead to a conclusion that Chatbot is more efficient than human in most cases. Twenty-five percent of customer service and customer support operations will transversely integrate virtual customer assistant (VCA) or chatbot technology into engagement channels by 2020, compared to less than two percent in 2017, according to Gartner, Inc. Despite of this growing trend, the current applying rate in the logistics industry is much smaller. Therefore, using Chatbot means taking more advantages. And our new Chatbot system can handle demand in many cases. FPT Software has received requests from leading logistics providers in the US and France to make a new friendly automotive interactive system. In this case, the Chatbot system developed by FPT Software can help their clients to make inquiries about shipment automatically without human involvement.

The Solutions

A project to develop a new Chatbot was deployed by FPT Software including these functions:
  • Quote Inquiry: users can ask about estimated delivery cost, estimated delivery time, comparison between services‚
  • Shipment Tracking Query: users can track current location/status of their package, expected delivery time, delay event, and whether a signature is required.
  • User Manual: users can also be instructed to fill in forms, to use apps/website, payment‚

One of the advantages of this system is that it can self-improve through time by using built-in AI and also process various languages such as Japanese, English, French, Vietnamese, etc. Future enhancements could include multiple other scenarios, integration with multiple systems, support of voice chat, compatibility with multiple platforms, etc.  

Technologies Used:

  • Processing natural language and configuring conversation logic:
    • In the PoC, IBM Watson Conversation is adopted for short chats including 3-5 sentences.
    • For future development, tentatively IBM Watson Knowledge Studio and IBM Watson Alchemy Language will be involved for longer conversations.
  • Web server: Developed in Node.JS to handle all requests which currently allows chatting from web browser and mobile app. The coming features would include interactions with Facebook, Twitter and Hangout.
  • Web client: HTML5, CSS3, JQuery 

The Benefits

  • The new Chatbot system can reduce workforce by 70% and free-up staff for higher value tasks.
  • It communicates with customers faster, 24/7 with accurate information about shipments.
  • The Chatbot is also friendly, intelligent, which allows users to interact using natural language without any prerequisite training.
  • Overall, it improves service quality and gradually replaces traditional platforms of communication.