<p>This study investigates the effect of Bacillus subtilis bacteria on the compressive strength of concrete by incorporating five different bacterial concentrations: 10³, 10⁵, 10⁷,10⁸, and 10⁹ cells/ml. Ten input parameters Cement (PPC), Fine Aggregates, Coarse Aggregates, Fly ash, Calcium Lactate, pH, W/C, Bacterial concentration, Temperature of curing tank, and Humidity used to determine CS of self-healing concrete. Experimental results revealed that the addition of Bacillus Subtilis concentration of 10⁷ and 10⁸ cells/ml has high potential of enhancing compressive strength as compared to nominal concrete mix. Self-Healing efficiency of the degraded concrete was increase to 99% compared to the control mix in 10⁷ and 10⁸ Bacillus Subtilis concentrated concrete specimens. Three machine learning techniques Random Forest, Random Tree and ANN models were applied and Random Tree model outperformed among three models on bacterial concrete with CC values 0.9743 and 0.87799 for training and testing data sets respectively. Specifically, the interrelationship amongst significant input parameters is established with the development of a mathematical model by using SPSS. From the sensitivity analysis, it was found that Bacterial concentration was the most significant input parameter.</p> Graphical abstract <p></p>