Data-driven decision making for engineers

Summary

Use guided Python notebooks to analyse engineering data, make defensible comparisons, quantify uncertainty, build simple predictive models and avoid common statistical traps in decision-making.

Engineers are increasingly expected to justify decisions using evidence from test data, sensor data, maintenance records and operational performance data. Many decisions, however, are still made using averages, trends or comparisons that do not properly account for variation, uncertainty or noise.

This course uses guided Python notebooks and engineering datasets to help delegates move from raw data to defensible engineering decisions. Python is used as the practical analysis environment, not taught as the main subject.

Engineers are increasingly expected to justify decisions using evidence from test data, sensor data, maintenance records and operational performance data. Many decisions, however, are still made using averages, trends or comparisons that do not properly account for variation, uncertainty or noise.

This course gives engineers the statistical judgement and practical methods needed to understand what their data is actually saying before drawing conclusions.

Who should attend?

How will I benefit?

After the course you will be able to: 

1. Summarise and visualise engineering data correctly before making decisions.
2. Distinguish meaningful patterns from noise, random variation and misleading averages.
3. Use hypothesis testing and confidence intervals to compare processes, outcomes and measurements.
4. Quantify uncertainty in engineering conclusions and predictions.

Use guided Python notebooks to visualise data, compare engineering outcomes, quantify uncertainty and build simple regression models.

Key topics

• Python notebooks for engineering data analysis
• Data cleaning, preparation and quality checks
• Descriptive statistics for engineering datasets
• Data visualisation, distributions and outlier detection
• Understanding variation, noise and uncertainty
• Hypothesis testing and confidence intervals
• Comparing processes, outcomes and test results
• Correlation, regression and model evaluation
• Common errors in engineering data interpretation
• Worked Python-based engineering case study
  
   

Meet our trainers

These trainers regularly teach Data-driven decision making for engineers.