Story Title: Build Machine Learning Model to Predict REQ Age

 

Story Title: Build Machine Learning Model to Predict REQ Age

User Story

As a data analyst, I want to build a machine learning model to predict the REQ Age (Request Age), so that I can provide insights on the time taken to close requests and improve our service efficiency.

High-Level Acceptance Criteria

1. Data Collection:

   • Gather historical data, including features such as Catalog Item, REQ Short Desc, REQ Age, RITM Short Desc, RITM Age, Business Sector, Region, and FTE.

2. Data Preprocessing:

   • Ensure data is cleaned and preprocessed, including handling missing values, normalizing numerical features, and encoding categorical variables.

3. Model Selection:

   • Evaluate multiple machine learning algorithms such as Linear Regression, Random Forest, Gradient Boosting, and Neural Networks to identify the best model.

4. Model Training and Evaluation:

   • Train the selected model using historical data and evaluate its performance using metrics such as Root Mean Square Error (RMSE) and Mean Absolute Error (MAE).

5. Feature Importance Analysis:

   • Analyze the importance of different features in predicting REQ Age to gain insights into key drivers of request completion time.

6. Model Deployment:

   • Deploy the trained model into the production environment, ensuring it integrates seamlessly with existing systems.

7. Documentation and Reporting:

   • Document the model development process, including data preprocessing steps, model selection rationale, and evaluation metrics.

   • Create a report summarizing the model performance and key findings.

8. Monitoring and Maintenance:

   • Set up mechanisms to monitor the model's performance over time and retrain it periodically as new data becomes available.