1. Failed to validate; converge to different peaks (EM alg. converge to local optimum) 2. synthetic data fitting; (w.o. using extended kalman filter) 3. - service capacity, then compared to the demand you're getting leads to the sales you get in the store and part of it is kind of sales in terms of the number of people you serve versus how much each person buys and. - reward that paper, so this is that's the goal, so what you have seen in that paper is very similar to what we want to do, the problem with that paper, the way it was is that that estimation passenger essentially broke down. - 00:04:30Because, because when I was doing validation testing, so you can estimate anything and then hope that it is working, but you don't know whether it is working. - 00:04:43So when I created some synthetic data using the model as the engine yeah and then try to estimate the parameters of the model, using the method, I was using I wasn't getting reliable estimates and optimization algorithm for estimation converge two different local peaks.