My answer (as a mathematician by background who lectures in engineering) to why engineers need a good grounding in applied probability and statistics is "All engineers deal with data, and all engineers deal with reliability."
There are of course other purposes for which probability and stats are important, but I think that one sentence illustrates why probability and statistics is a core engineering discipline - indeed I frequently argue that it is hard to find a topic which is traditionally thought of as "engineering" which has such wide importance across all engineering disciplines.
Chris
--
Dr. Chris Dent,
Lecturer in Energy Systems Modelling,
School of Engineering and Computing Sciences,
Durham University,
South Road,
Durham DH1 3LE. U.K.
Tel: +44 (0) 191 33 42451
Personal website: http://www.dur.ac.uk/ecs/profiles/?id=7876/
Durham Energy Institute: http://www.dur.ac.uk/dei/
Postgraduate Mentor at University College: http://www.dur.ac.uk/castle.mcr/
________________________________________
From: A UK-based worldwide e-mail broadcast system mailing list [[log in to unmask]] on behalf of Dan Grove [[log in to unmask]]
Sent: 17 June 2013 15:21
To: [log in to unmask]
Subject: Why engineers need statistics
Hi (anyone following this thread )
Here's what we say when we introduce the Bradford University course called Advanced Statistics for Engineers.
A key success criterion for an engineered system is minimal variation in functional performance i.e. it does what it is supposed to do every time;
For any engineered system, variability in the input and in the system itself will determine variability of the output - statistical methods allow engineers to understand and measure variation of all kinds;
Statistical methods (including designed experiments) help engineers to develop models of complex systems;
Statistical methods guide engineers in using data to make decisions for product and process design, for process control and for problem solving.
Re the material from Dr Nic (Nicola Petty): I wouldn't want to use the video 'Experimental Design Elements' for engineers' - too much emphasis on randomisation! There are 2 issues, effectiveness and practicality.
Effectiveness: in a simple comparative experiment like the video example, we can expect randomisation to deal with the lurking variables (provided we have plenty of units); not so in many industrial experiments where we try to measure almost as many effects as there are units (runs). In this situation it is likely that a lurking variable will be highly correlated with one or more effects, for any run order. I prefer to get engineers to think about what might be lurking, and design the experiment accordingly.
Practicality: in most industrial experiments randomisation comes with a cost - I would rather have the experiment run in a 'convenience' order than not at all.
Dan Grove
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|