The current system for inventory management consists of enterprise resource planning which solves the problem of asset visibility but the quantum of inventory held at each echelon and warehousing still required manual intervention. We would like to automate the process using AI which will ensure correct inventory be placed at appropriate regiment in the appropriate time reducing time required for inventory management, ensuring availability and avoiding over provisioning of the resources.

What it does

Our AI model will predict the amount of resource that will be consumed by each regiment the next month. This value will be tallied with the currently available resources in each regiment and show what amount of which resource must be supplied to which regiment to avoid scarcity or over provisioning of any resource.

How we built it

We have created a database containing 4 tables. - First Table: To store current available stock for each item in each regiment. - Second Table: To store the current amount of each item available in the Warehouse. - Third Table: To store the minimum quantity of each item that should be available at each regiment. - Fourth Table: To store the login credentials of the authorized users. We have encrypted Our database data with RSA Encryption.

Next Step is to build an AI model. We have created the model using Keras Functional API.

UI It will display the current availability of each item in each regiment, and stocks available in Warehouse. It will also display the quantity of each item that should be transferred from warehouse to the regiment.

Challenges we ran into

Connecting the model with database for real time updates.

Accomplishments that we're proud of

Predicting the Monthly Consumption at each location precisely.

What we learned

Working with different modules and compiling them altogether.

What's next for AI in Inventory Management

Training the model continuously.

Share this project: