Pulp: Setting Constraints that are not lpSum based but rather in a range?

Question

I am trying to understand the pulp library in python, specifically in defining constraints that don't involve lpSum and are just straight bounds on one of the variables.

orders_dict_test = {"Orders": [11, 22, 33, 44, 55, 66, 77],
                    "Num_Days": [1, 2, 2, 3, 3, 5, 6]}

How would one create a pulp constraint using "Num_Days", where we only want to include "Orders" where "Num_Days" is >= 1 and <= 4?

I am just having a hard time understanding constraints that doesn't involve a lpSum as many of the examples I have seen are focused on that.

Any resources or help would be great.

Thank you.

Answer 1

I don't fully understand the question, or what you are trying to achieve. As a general point though when creating constraints there tend to be two things you can do with 'sets' of variables (a set is just a collection of variables which you could define in any way you like).

You can either sum over that set of variables, or you can apply a constraint for each of the variables in that set.

So for your example it would be quite straightforward to create the subset of variables which satisfy orders_dict_test['Num_Days'] >= 1 and <= 4 , and you could then sum over that subset or apply a constraint for each variables in that subset.

The code below will create a list of tuples of ('orders', 'ndays') which satisfy that constraint. Extracting the actual variables will depend on how you have defined them. I would suggest that you probably want a unique key of some description.

orders_dict_test = {"Orders": [11, 22, 33, 44, 55, 66, 77],
                    "Num_Days": [1, 2, 2, 3, 3, 5, 6]}

print([(o, nd) for o, nd in zip(orders_dict_test['Orders'], orders_dict_test['Num_Days']) if nd >= 1 and nd<= 4])