Why Engineers Can’t Skip Statistics
3 Reasons You Need It
Knowing statistics is incredibly beneficial for engineers.
Statistics is a branch of mathematics that deals with the collection, analysis, interpretation, presentation, and organization of data. It can be useful for engineers in various ways, such as:
- data analysis,
- quality control, and
- decision making.
Data Analysis
As an example, let’s say we work for a propane tank manufacturer and we’re designing a new propane tank. We currently have two different designs, and we want to know which design is better. The thing we want to know is: which design can hold more propane before failing? Let’s suppose that we run the experiment and find the following average results:
- Design A: 155 psi
- Design B: 160 psi
Design B holds more pressure, so Design B is obviously better, right?
Not so fast. With only a 5 psi difference, it’s hard to tell, without more information. Within each design, how much does the maximum pressure vary between each of the 10 samples?
Let’s suppose that Tab. 1 contains the raw experimental data. You can see that the averages are the same as the reported values above, but the standard deviations are very different.
From the experimental data in Tab. 1, we fit distributions, as seen in Fig. 2. From this you can visually see that Design B is not necessarily better, as Design B varies significantly more. Sure, there are many more Design B tanks that can hold pressures much higher than Design A tanks. But in this case, we also see that quite a few Design B tanks will fail at pressures well before most of the Design A tanks. And this could be a huge safety hazard! So we would likely want to go with Design A, since the tank pressure performance is “more predictable”.
The above is a simplified example of data analysis that we could run. But there are countless other examples of analyses we could run. Some are just as simple, others are more complex.
Quality Control
Statistics can be used to monitor and control the quality of a process or product, by identifying trends and patterns in the data that may indicate a problem. This can help engineers to identify and fix issues before they cause problems.
As an example, suppose that our propane tank company is manufacturing a valve for our propane tank. This valve requires a certain level of precision in its dimensions. The company wants to ensure that the valve meets the required specification, and that the manufacturing process is consistent and reliable.
To do this, we might take regular samples of the finished product and measure its dimensions. These measurements can be collected and analyzed using statistical techniques to monitor the quality of the manufacturing process.
For example, we might calculate statistical indicators such as the mean, median, and standard deviation of the measurements. If the mean or median is consistently far away from the intended mean or median, this could indicate a problem with the process. Similarly, a large standard deviation could indicate that the process is not consistent, and that the valve dimensions are varying significantly from one batch to another.
By analyzing the data using statistical techniques, we can identify trends and patterns that may indicate a problem with the manufacturing process. This can help them to take corrective action and improve the quality of the product.
This is just one example of how statistics can be used for quality control. There are many other statistical techniques and indicators that can be used for this purpose, depending on the specific requirements and characteristics of the process or product.
Decision Making
Statistics can help engineers to make informed decisions by providing a systematic way to analyze and interpret data. This can be useful in areas such as risk assessment and resource allocation.
Let’s again suppose that our propane tank company is considering two different propane tank designs. Design C has a higher upfront cost, but is expected to have a higher return on investment (ROI) over the long term. Design D has a lower upfront cost, but is expected to have a lower ROI over the long term.
To help decide which option to choose, we might use statistical analysis to evaluate the expected ROI for each option. We might use techniques such as Monte Carlo simulation to generate a range of possible outcomes for each option, based on assumptions about market conditions and other factors.
By analyzing the results of the Monte Carlo simulation, we can determine the likelihood of different ROI outcomes for each option. This can help them to make an informed decision about which option is more likely to be the most profitable in the long term.
This is just one example of how statistics can be used for decision making. There are many other statistical techniques that can be used to evaluate different options and make informed decisions, depending on the specific context and requirements.
Conclusion
Effective engineers use data analysis to analyze and interpret data, and draw meaningful conclusions from it. They also conduct quality control by identifying trends and patterns in data that may indicate a problem. Lastly, they use statistics to evaluate different options and make informed decisions.
Further Reading?
Interested in more engineering topics?
Want to see what a Silicon Valley Engineer does all day?
