If machine A and B are of different capacity, but built in same year, what is the model that we should use?

The question seems to be incomplete as it doesn't specify what kind of model is being referred to. However, if we're talking about a statistical model to compare the capacities of the two machines, you might consider using a two-sample t-test if the capacities are normally distributed. This test compares the means of two independent groups in order to determine whether there is statistical evidence that the associated population means are significantly different.

If the data is not normally distributed, a non-parametric test like the Mann-Whitney U test could be used.

If you're referring to a type of machine learning model to predict some outcome based on the features of the machine (including capacity and year built), you might consider multiple linear regression if the outcome is continuous or logistic regression if the outcome is binary.

A decision tree model could also be used as it can handle both numerical and categorical data and doesn't make any assumptions about the distributions of the predictors.

Remember, the choice of model also depends on other factors such as the size and quality of your dataset, the presence of outliers, and the relationships between variables.