As commonly assumed in scheduling problems involving earliness measures, we consider here only non-delay schedules. (Otherwise, optimality is trivially obtained by sufficiently delaying the jobs.) Such schedules (with no idle times prior to the first job and between consecutive jobs) are justified in many manufacturing systems, where the production process cannot be stopped until the entire set of jobs is finished.We prove that the problem studied here is NP-hard. Then, we propose a pseudo-polynomial dynamic programming (DP) algorithm, implying that the problem is NP-hard in the ordinary sense. Our numerical tests indicate that the proposed DP is very efficient, and the solution of problems of medium size (up to 200 jobs) requires very reasonable computational effort.