It has been daunting for the warehouse staffoand top-level management of the company to counter operational challenges with their existing supply chain management tool. They faced difficulties extracting information related to incoming orders, previous sales, and inventory status. In addition, the staff were dependent on IT teams to access the dashboard to retrieve information and implement changes wherever required.
Our team of experts investigated how data is stored, structured, and retrieved from the warehouse system. The tool was working well, but it was incapable of bringing consolidated data related to items in stock, pending, dispatched, and delivered orders. We wanted to build a free-flow chat that provides insights into the warehouse and sub-houses in minutes.
Warehouse staff struggled to access critical data, including shipments, inventory, and sales history, leading to wasted time and operational slowdowns.
It was required to detangle the complex algorithms that are run repeatedly until they find a desired result. For example, there is frequent inaccurate demand forecasting and a communication gap between partners.
SQLCoder-34B, an open-source Language Model (LLM), was explored due to privacy concerns around ChatGPT4. However, it faced challenges related to working and performance. It could only answer 2 out of 10 questions.
We have created a robust platform for warehouse personnel by blending ChatGPT4 with open-source LLM. It has helped thousands of people working within the company. Approximately a list of 40,000 items is brought into a single chat to solve users’ queries. The strategic fusion of ChatGPT4 and LLM exceeds the potential of the company
