“…Since the experiments last limited times, we may assume that t 0 equal the elapsed time t e of the experiments. The results of data fittings available in the literature [5,48,49,76,87,92,94,129,133,134,137,145] reveal that the relaxation times form two groups: relaxation times less than the elapsed time τ i < t e ⇒ τ i /t e < 1 and relaxation time greater than the elapsed time τ i > t e ⇒ τ i /t e > 1. Consequently, there are two groups of fractional orders α i , namely: (i) For τ i /t e < 1, we have α i ∈ [0.5, 1] related to fast relaxations and (ii) for τ i /t e > 1, we have α i ∈ (0.5, 1] modelling slow relaxations.…”