Management of Technology (Application oriented)

1. Course Title:

Predictive Analytics and Industry Applications

2. Language of Instruction:

English

3. Course Code:

INT647020

4. Credits:

3

5. Course Objectives:

This course is designed to equip learners with predictive modeling techniques (time series & cross-section), model evaluation and deployment, and domain-specific applications (manufacturing, logistics, marketing, finance, healthcare): • Provide students with solid foundations in predictive modeling methods (regression, classification, tree ensembles, gradient boosting, neural nets, time-series forecasting), including their assumptions, strengths, and limitations. • Teach end-to-end predictive workflows: problem scoping → data engineering & feature engineering → model building → validation & calibration → deployment & monitoring (basic MLOps). • Develop the ability to select appropriate predictive techniques for industry problems and to translate model outputs into managerially useful decisions and KPIs. • Give hands-on experience with mainstream tools (Python: pandas, scikit-learn, XGBoost, TensorFlow/Keras; R: tidymodels/caret; plus MLflow for lifecycle) and time-series libraries (e.g., Hyndman’s forecasting tools / statsmodels). • Sensitize students to ethical issues, bias, explainability and regulatory constraints when applying predictive models in real sectors.

6. Brief Description of Course Content:

This applied course combines theory and practice of predictive analytics for industry. Students learn statistical and machine-learning models for forecasting and classification, feature engineering, model selection and hyper-parameter tuning, cross-validation, model interpretability and monitoring. Time-series techniques (ARIMA, exponential smoothing, state-space models, Prophet), and modern ML approaches (tree-based ensembles, gradient boosting, neural networks) are covered. The module includes hands-on labs and one team project (industry dataset) where students deliver a predictive solution, a dashboard, and an implementation/monitoring plan.