The advantage is we can have a variable whose value is safeguarded against reassignment..i.e. helps us to define a constant value for the method. We can do it globally but if it is only required for a particular method it is better to put it in local scope.
Eg: We have a method to calculate the area of circle, the constant pi could be local final variable. By making it final we ensure its value is constant
while variables can change value (because they are variable), it is generally bad practice to reuse variables.
If you obtain a value for a local variable and that value is not expected to change, you can prevent bugs by making the variable final.