Comments on ‘Competitive analysis of a better on‐line algorithm to minimize total completion time on a single‐machine’

For the single machine scheduling problem of minimizing the total completion time, Torres (2003) presented a semi‐online algorithm under the assumption that release dates are known in advance, and showed that it was $\sqrt{3}$ ‐competitive. However, there are flaws in the proof, and the conclusion about the competitive ratio is not correct. In this note, we show that the semi‐online algorithm cannot perform better than the best non‐clairvoyant online algorithm with a competitive ratio of 2.