Abstract
If you were running a PV factory, what would be most important to you in terms of that factory's performance? Most people would say the average efficiency, and for the next few years they may well be correct. But in many stable, high-value, manufacturing industries, the notion of quality is paramount, and that is quality defined in the more general sense of making the same thing every time, consistently. The direct measure of your ability to make something the same every time is contained in the variance. Ideally, if you can work out the sources of this variance, then you have an even more sensitive indicator of this quality and you will also know what you need to do to improve it.
This seminar will introduce a statistical modelling technique known as Path Modelling, a type of Structural Equation Modelling (SEM), and show its utility in helping to understand the sources of variance in cell manufacturing data. A path model is an expression of the relationships between the measured data and the causes of these relationships. It is a versatile and powerful technique, it can work to determine the influence of what is known as latent or unmeasured variables from the relationships in the measured ones. Path Models can work successfully using only physical models of the cell and standard end-of-line measurements. But the more data that is known or measured about a cell, the more sophisticated and complete the model can become. This technique therefore also becomes a method of assessing which types of metrology are most effective in helping to understand variance and so maintain a stable production.
Click here to see all available video seminars.
Click here to go to the SPREE HOMEPAGE.
Brief Bio
Rhett Evans received his Bachelor of Engineering degree (Electrical) with Honours Class 1 from UNSW in 1997. Rhett worked with Pacific Solar and then CSG Solar in the development of thin film technologies, working initially in process development, then as the Pilot Line manager and finally as the Technical Director. Rhett was also the Technical Director of Suntech R&D Australia until 2014, working with the manufacturing facilities of Suntech in China. Rhett began a PhD at UNSW in mid 2013, and his area of research is in developing statistical models to characterise PV manufacturing, and using data driven approaches to optimise PV manufacturing around product principles. Since 2014 Rhett has also been a director of Solinno Pty Ltd, working as a consultant in analytics projects in the photovoltaic industry, mostly with manufacturers.
