AI Chatbot for Customer Support
in Logistics
About the Client
This article describes a real-life project. However, we cannot disclose our client’s name & the project details for privacy purposes.
A logistics solution provider in Vietnam faces increasing customer service challenges as demand for real-time support grows. While the company deploys a rule-based chatbot across multiple channels such as Zalo and ChatPlus, the system is limited to keyword detection and basic message distribution. As a result, most inquiries still require manual handling from staff across Sales, Technical, and Logistics departments.
The company aims to implement an AI-powered chatbot to automate customer query resolution, reduce response delays, and provide accurate request routing at scale.
Project Overview
The project focuses on developing an AI-powered system that uses in-store photos to automatically analyse product displays and shelf arrangements. By replacing manual inspections with automated image recognition, the client aims to streamline daily store management and improve overall accuracy.
Industry
Logistics
Technology
AWS, Generative AI, Google Gemini, Vector DB
Country
Vietnam
Timeline
May 2025 – Aug 2025
Challenges
The company faces significant difficulties in managing customer inquiries effectively across multiple channels. Despite having a basic chatbot in place, the system fails to meet growing demands for speed, accuracy, and scalability. Key challenges include:
Multi-channel support dependency
Customer inquiries come through platforms like Zalo and ChatPlus, but most interactions still require direct responses from employees across Sales, Technical, and Logistics departments.
Limitations of the existing chatbot
The rule- and keyword-based chatbot works mainly as a message distribution tool, handling only a few simple scenarios such as outside working hours or new customer greetings.
Low accuracy and inefficiency
Relying solely on keywords often leads to misrouted messages and unanswered questions. Employees spend extra time transferring chats between departments and responding to repetitive inquiries.
Slow customer response time
With staff overloaded, customers frequently experience long waiting times for replies, reducing overall satisfaction.
Need for an advanced solution
To resolve these issues, the company requires an AI-powered chatbot that clarifies customer questions, ensures accurate routing, and directly responds to queries within a defined scope.
Solution
To address these challenges, Rikkeisoft builds an AI-powered Chatbot designed to seamlessly handle customer inquiries, route them to the right departments, and provide direct answers when possible.
Key solution features include:
Unified message handling
The chatbot automatically receives all incoming messages from customers across every platform. For unresolved requests, it engages in follow-up questions to identify and classify the inquiry based on predefined topics.
Clarification & categorization
The chatbot asks additional questions to gather the necessary details related to each topic, ensuring accurate context before taking action.
Smart response & routing
Depending on the request, the chatbot either replies directly within its scope of knowledge or routes the message to the responsible department. Along with the transfer, it also provides extracted information to minimize repeated clarification work.
Human-like communication
Responses are generated to mimic the style and tone of real support staff, taking into account factors such as form of address, timing, names, and level of familiarity. This ensures replies are natural, context-aware, and flexible like those of a human agent.
Integration with internal resources
The chatbot connects to internal systems via APIs, accessing customer databases, process documentation, past chat history, and employee response records to deliver accurate and personalized support.
High performance & scalability
Designed for low latency and high load tolerance, the chatbot supports tens of thousands of customers simultaneously without compromising response time.
Results
60%
reduction in customer response time
30%
of inquiries fully resolved by the chatbot
95%
routing accuracy
50%
reduction in employee workload
The implementation delivers measurable improvements in both efficiency and customer experience. The system now supports over 300 types of ID documents across multiple Asian countries and provides robust multilingual recognition. Document processing time reduces by 80%, significantly accelerating customer onboarding, while operational costs fall by 70% thanks to the elimination of manual data entry. Most importantly, the client achieves a more secure, reliable, and user-friendly verification process, strengthening customer trust and improving overall satisfaction.