Foundations of Data-Driven Thinking 2.5 credits

**Unlock the value of your manufacturing data with the *Foundations of Data-Driven Thinking*.**
This course targets a wide range of roles in industry and the public sector, such as decision-makers, engineering and technical specialists in design and production, and R&D personnel, who seek practical insight into how data-driven AI can be leveraged to create value in their organizations. Working hands-on with real industrial datasets, you’ll learn how to handle common manufacturing data types, such as sensor and time-series data, process and production logs, and quality and test measurements. You will apply methods to: - assess data quality and readiness (noise, drift, missing values, inconsistencies) - clean and prepare data for analysis - perform exploratory data analysis to reveal trends, relationships, and root causes - visualize and communicate insights in clear, decision-oriented ways. You’ll also learn how to use the gained insights to plan the next step in an analytics or AI initiative, such as building solutions for predictive maintenance, automated quality inspection, anomaly detection, scrap and yield improvement, process monitoring, or throughput and cycle-time prediction.

The course introduces foundational concepts of data-driven thinking with a focus on industrial manufacturing applications. Emphasis is placed on understanding data types, data quality, and exploratory data analysis (EDA) as a basis for sound analysis and informed decision-making. Participants work with real or realistic industrial data to develop practical skills in data handling, visualization, and interpretation. The course highlights how early data quality assessment and exploration support robust process understanding and provide a foundation for later use of advanced statistical and AI-based methods. The course consists of the following elements: - Foundations of data-driven decision-making in manufacturing. - Manufacturing data types, including numerical and categorical data. - Data quality, data cleaning and pre-processing. - Exploratory data analysis (EDA): univariate and multivariate EDA methods - A glimpse into modern AI methods for modelling and prediction.