I have a question about constructing statistical control charts. Suppose I have an attribute- say, "weight," and I'd like to track measurements of this weight for several different drums, i.e. $$\text{drum}_1, \text{drum}_2, \text{drum}_3,...,\text{drum}_n.$$ The time interval for each drum is constant; assume $\Delta t =10$ minutes, where $t_0=0$ minutes.
However (this is where things aren't so clear to me), for each $\text{drum}_i,$ there is a different sample size for time- for example, $\text{drum}_1$ might only have data for times $t_0, t_1, t_2,$ and $t_3$, whereas $\text{drum}_6$ might have data for times $t_0,...,t_{20}.$ In other words, the time collection stops at different points for each drum.
Would anyone know how I could properly set up $one$ control chart that would display all drum data? I assume I would have to normalize something, but I'm not sure what, and how I would go about doing so. Perhaps I would normalize the time?