Abstract The purpose of this paper is to establish some guidelines for designing effective Estimation of Distribution Algorithms (EDAs). These guidelines aim at balancing intensification and diversification in EDAs. Most EDAs are able to maintain some important linkages among variables. This advantage, however, may lead to the premature convergence of EDAs since the probabilistic models no longer generating diversified solutions. In addition, overfitting might occure in EDAs. This paper proposes guidelines based on the convergence speed analysis of EDAs under different computational times for designing effective EDA algorithms. The major ideas are to increase the population diversity gradually and by hybridizing EDAs with other meta-heuristics. Using these guidelines, this research further proposes an adaptive EA/G and EA/G-GA to improve the performance of EA/G. The proposed algorithm solved the single machine scheduling problems with earliness/tardiness cost in a just-in-time scheduling environment. The experimental results indicated that the Adaptive EA/G and EA/G-GA outperform ACGA and EA/G statistically significant with different stopping criteria. When it comes to the intensification of EDAs, heuristic method is combined with EDAs.