Skills: Programming, Data Science, Timeseries

Tools: Python, Optuna, XGboost, sklearn, ARIMA

Keywords: Python, Time Series Analysis, XG Boost, Optuna, CPG Industry, Manufacturing Process, Process Optimization, Energy Conservation

I was a part of the development of a robust Manufacturing Pipeline Optimization specifically designed for the Consumer Packaged Goods (CPG) industry. This project aimed to minimize energy consumption and optimize raw material usage while ensuring that the end-product was of optimal quality.

The Manufacturing Pipeline Optimization project was a collaborative effort involving a team of three individuals. Its primary goal was to enhance the manufacturing process within the Consumer Packaged Goods industry. The project focused on minimizing energy consumption and optimizing raw material utilization while maintaining high product quality. To achieve this, the team developed a robust optimization strategy, incorporating process data checkpoints, a detailed process overview, and controlled variables. Optimal value ranges for each metric were established to facilitate identification of deviations from ideal conditions. By implementing these optimization measures, the project aimed to achieve significant efficiency improvements and cost savings for the manufacturing pipeline while ensuring the end products met the defined quality metrics.

Data filtering and wrangling

EDA on Time series data

Time Series Forecasting

Explainable Components in AI models for Feature Importance understanding on each batch, each process, each hour, day and week.

Using an RL agent to accomplish required goals

Team Size: 3

Process Data - Checkpoint or quality check data with appropriate time stamps

Process Description - Overview of the process that a single batch passes through

Controllable Variables Set- What Additional variables are being controlled

Optimal Value ranges for each metric

End Product Quality definition metrics