This is fairly common, as reaction rates (and rates of change in general) tend to be high at the start of a batch or fed batch simulation and decrease with time (e.g. as substrate is reacted). Some common causes are in mass transfer, where initially the system is far from equilibrium, so the rate of mass transfer is high, resulting in small timesteps. Poor specification of initial conditions can also cause this (e.g. in a vapor-liquid phase system, specifying the water is in the liquid phase at 110 C and 1 bar, the water will all transfer rapidly to the vapor phase, causing the run to start slowly).

Open the event log to monitor the progress of the solver and check which variables are controlling the stepsize.

Show me other factors affecting run time.

This is probably due to a large number of equations being solved. Check the number of equations in the solver properties window. If this is greater than 15, consider simplifying the model (i.e. remove mechanisms) to speed up the run. Open the event log to monitor the progress of the solver and check which variables are controlling the stepsize.

Another source of slowness is if there are a wide range of rates in the model. It may help to simplify the model (reduce the number of equations) or try a different solver (e.g. Backwards Euler).

In some cases, very large model parameter values (e.g. rate constants, mass transfer coefficients) may be set to enforce equilibrium. It may be possible to reduce these parameters without affecting the solution, which may reduce the run times.

Show me other factors affecting run time.