A Practical approach for exponentiation of QED corrections in arbitrary processes