0
$\begingroup$

I have a general inquiry regarding a project I am working on. I cannot reveal too much, but I would like to gauge the community here and hopefully be pointed towards the right direction in terms of what to learn, what to research, and what to look into with respect to my problem.

I am trying to forecast several artificial processes that have only been in production for several months to a year. The processes themselves are not in steady state and we are experiencing high dispersion in the parameter that is to be forecasted.

Since the process is not in steady state, I am thinking of applying a simple moving average to put more weight on recent data to create my forecast line.

My main issue comes with process control. I am not too sure how to go about it. I am thinking of constructing an upper threshold limit that will raise flags for those monitoring the process if crossed. Ideally, I would want to be able to use the forecasting model and this upper limit value and superimpose live data on top of it to make live adjustments to the process.

I am just simply looking to be pointed in the right direction of where to learn/find the information and tools I need for this. Thanks.

$\endgroup$
0
$\begingroup$

First build a forecasting model using moving average or neural network.

Second use a model predictive controller (MPC) which is an optimization based controller that can have your forecasting model plus any constraints you want on the variables (upper threshold limit).

Hope this help.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.