Hi,
We are trying to create optimizer where we have the multiple decision variables like deciding on the number of line runs, forecast number to be allocated on a line and inventory norms to be met for a particular product . for instance my objective equation is as below:
ax+by+cz
there are multiple objectives which has to be met with priority orders (without weightages). How can we use Anaplan’s Optimizer to configure this?
there are two possible approaches we could think of:
Also, unit of measure of all decision variables in objective function may not be same. For instance – x is in units - units of a product to be produced , y are in days- days of inventory cover to be maintained and z are in number- number of line runs recommended.
how can we configure such a scenario in Anaplan's optimizer
Any help on this would be really appreciated
That's an interesting one. As far as I know Anaplan Optimizer will have only one Objective as an output. You can have as many variables or constraints in the model but as an output Anaplan will let you choose only one. Also Anaplan doesn't have an option to prioritize the constraints
However it will be great if you can explain the requirement by calling out -
What are your Variables and the calculations?
What are your Constraints and what is the equation?
Objectives - Minimize or maximize?
Thanks,
Misbah
Hi Misbah,
we are trying to minimize number of runs , maximize the demand which can be scheduled on a factory line, trying to meet the norms to the maximum extent we can.
Now all these go in priority, my first priority would be to meet the demand , then try to meet norms, then once all this is met try to minimize the number of runs that a line should take to minimize the changeover time.
As I mentioned earlier Anaplan optimizer will work only on one objective at a time. Let's say in your case you need to
1. Maximizing the demand
2. Minimizing the number of runs or Optimizing the production Run rate
In that order.
You will have to take each piece separately and work on it. Not sure how that will pan out to be because Once you have run the first optimizer, output of that has to be an input or constraint to the second optimizer. My understanding is that you can't club these two objectives together.