Failure modes are classified by their type and then their probability of occurrence. The type may correlated with their effect on the performance. So, from my point of view, it is the degradation rate per one time occurring times the occurrence rate which can be used as a measure for seriousness of the failure mode. Here the degradation rate can be measured by decrease in the solar cell efficiency for only one event occurrence of the failure mode.
Thank you very much Dr. Zekry. The main problem is, if a single module have different failure modes than how one can estimate the degradation rate for a particular failure mode.
You have asked for how to rank failure modes with a severity rating, and this implies you want to do something about them. There are several conditions - one is the panel is being developed and you don't want to design in defects. Secondly the panel is being manufactured and you want to cut the reject rate. Don't forget that for every step down the manufacturing process the cost to fix a fault increases massively, maybe ten times each step. Finally the panel might be installed and you want to know what failures are most likely so you can be ready to fix them. For example, the cost to fix a short circuit might be cents when you're stringing the cells together; it might be dollars once you encapsulated the panel; it might be thousands of dollars once it's installed in some remote location. So the severity ranking will change with circumstances. Personally I would rate the severity of failure purely by the cost to fix it, and that changes massively under varying conditions.
Thank you Dr Madan, Dr. Mourad and Dr. Maine for your comments. Yes Severity of failure can be analyzed by the cost to fixed it. For customer point of view we can also analyzed based on safety and power degradation.
The failure rate for certain failure mechanism is determined by accelerated tests using environmental testing chambers. They can be obtained also through long field observations, may be extending to the whole life of the solar panels. Because the statistical nature of the failure, one can apply quantitatively the results from certain type of panel to others. Please for more information follow the link: