See more projects
nda

The customer chose to protect their confidential information

Smart Warehouse Management System

About the client

The European division of a South Korea-based automotive manufacturer.

Location:South Korea
[object Object] on the map

About the project

The customer engaged Andersen to optimize its physical warehouse operations. For that, they asked our team to build an intelligent and feature-rich warehouse management system.

TEAM SIZE:11 specialists
Duration:8 months
Technologies:
Python
LangChain
Llama 3
Qwen 2.5 Coder
Milvus
CrewAI Agents
AI Models & Frameworks: Llama 3, Qwen 2.5 Coder
AI Tooling & Frameworks: LangChain, CrewAI Agents, Langfuse
Other: Python, scikit-learn, Milvus
Intelligent WMS architecture diagram
Intelligent WMS architecture diagram

Challenges

The customer faced major challenges in aligning supply with demand, especially in forecasting parts inventory stored in the data warehouse. The key issues that the requested warehouse management system software was to expected to resolve included:

  • Inaccurate demand predictions;
  • Delayed identification of stock shortages or surpluses;
  • Poor communication between planning and warehouse teams.

Solution

The resulting warehouse management system retrieves data from the database for analysis and forecasting. AI models (LSTM and SARIMA) are leveraged to perform time series forecasting for each inventory part. If forecasted data indicates critical issues (e.g. shortages, overstock), an alarm is raised for the customer’s warehouse inventory management staff.

The key delivered functionalities and elements, among others, are as follows:

  • Accurate and timely forecasting and modeling of supply and demand for each individual part;
  • A dedicated forecasting database for ongoing reference and analysis;
  • Real-time notifications for critical inventory events and anomalies;
  • The use of existing SQL queries to provide answers to user queries;
  • An intelligent chatbot to answer employees’ queries related to current and forecasted warehouse data;
  • Automated generation of SQL queries when no existing query is available;
  • Customizable dashboards tailored to specific needs and data preferences;
  • Observability and monitoring provided by a separate component of Langfuse.

Project results

Andersen's warehouse management application experts fully rose to the occasion and enabled total transparency and traceability, proactive and well-informed decision-making, and universal preparedness for the customer.

Let's talk about your IT needs

What happens next?

An expert contacts you after having analyzed your requirements;

If needed, we sign an NDA to ensure the highest privacy level;

We submit a comprehensive project proposal with estimates, timelines, CVs, etc.

Customers who trust us

SamsungVerivoxTUI

Let's talk about your IT needs